## What changes were proposed in this pull request?
This PR implements python wrappers for #13888 to convert old/new matrix columns in a DataFrame.
## How was this patch tested?
Doctest in python.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#13935 from yanboliang/spark-16242.
## What changes were proposed in this pull request?
jira: https://issues.apache.org/jira/browse/SPARK-16187
This is to provide conversion utils between old/new vector columns in a DataFrame. So users can use it to migrate their datasets and pipelines manually.
## How was this patch tested?
java and scala ut
Author: Yuhao Yang <yuhao.yang@intel.com>
Closes#13888 from hhbyyh/matComp.
## What changes were proposed in this pull request?
Just adjust the size of an array in line 58 so it does not cause an ArrayOutOfBoundsException in line 66.
## How was this patch tested?
Manual tests. I have recompiled the entire project with the fix, it has been built successfully and I have run the code, also with good results.
line 66: val yD = blas.ddot(trueWeights.length, x, 1, trueWeights, 1) + rnd.nextGaussian() * 0.1
crashes because trueWeights has length "nfeatures + 1" while "x" has length "features", and they should have the same length.
To fix this just make trueWeights be the same length as x.
I have recompiled the project with the change and it is working now:
[spark-1.6.1]$ spark-submit --master local[*] --class org.apache.spark.mllib.util.SVMDataGenerator mllib/target/spark-mllib_2.11-1.6.1.jar local /home/user/test
And it generates the data successfully now in the specified folder.
Author: José Antonio <joseanmunoz@gmail.com>
Closes#13895 from j4munoz/patch-2.
## What changes were proposed in this pull request?
model loading backward compatibility for ml.feature,
## How was this patch tested?
existing ut and manual test for loading 1.6 models.
Author: Yuhao Yang <yuhao.yang@intel.com>
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#13844 from hhbyyh/featureComp.
## What changes were proposed in this pull request?
jira: https://issues.apache.org/jira/browse/SPARK-16177
model loading backward compatibility for ml.regression
## How was this patch tested?
existing ut and manual test for loading 1.6 models.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#13879 from hhbyyh/regreComp.
## What changes were proposed in this pull request?
jira: https://issues.apache.org/jira/browse/SPARK-16130
model loading backward compatibility for ml.classfication.LogisticRegression
## How was this patch tested?
existing ut and manual test for loading old models.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#13841 from hhbyyh/lrcomp.
## What changes were proposed in this pull request?
Since we decided to switch spark.mllib package into maintenance mode in 2.0, it would be nice to update the package docs to reflect this change.
## How was this patch tested?
Manually checked generated APIs.
Author: Xiangrui Meng <meng@databricks.com>
Closes#13859 from mengxr/SPARK-16154.
## What changes were proposed in this pull request?
We recently deprecated setLabelCol in ChiSqSelectorModel (#13823):
~~~scala
/** group setParam */
Since("1.6.0")
deprecated("labelCol is not used by ChiSqSelectorModel.", "2.0.0")
def setLabelCol(value: String): this.type = set(labelCol, value)
~~~
This unfortunately hit a genjavadoc bug and broken doc generation. This is the generated Java code:
~~~java
/** group setParam */
public org.apache.spark.ml.feature.ChiSqSelectorModel setOutputCol (java.lang.String value) { throw new RuntimeException(); }
*
* deprecated labelCol is not used by ChiSqSelectorModel. Since 2.0.0.
*/
public org.apache.spark.ml.feature.ChiSqSelectorModel setLabelCol (java.lang.String value) { throw new RuntimeException(); }
~~~
Switching to multiline is a workaround.
Author: Xiangrui Meng <meng@databricks.com>
Closes#13855 from mengxr/SPARK-16153.
## What changes were proposed in this pull request?
`DefaultParamsReadable/Writable` are not user-facing. Only developers who implement `Transformer/Estimator` would use it. So this PR changes the annotation to `DeveloperApi`.
Author: Xiangrui Meng <meng@databricks.com>
Closes#13828 from mengxr/default-readable-should-be-developer-api.
[SPARK-14615](https://issues.apache.org/jira/browse/SPARK-14615) and #12627 changed `spark.ml` pipelines to use the new `ml.linalg` classes for `Vector`/`Matrix`. Some `Since` annotations for public methods/vals have not been updated accordingly to be `2.0.0`. This PR updates them.
## How was this patch tested?
Existing unit tests.
Author: Nick Pentreath <nickp@za.ibm.com>
Closes#13840 from MLnick/SPARK-16127-ml-linalg-since.
## What changes were proposed in this pull request?
Mark ml.classification algorithms as experimental to match Scala algorithms, update PyDoc for for thresholds on `LogisticRegression` to have same level of info as Scala, and enable mathjax for PyDoc.
## How was this patch tested?
Built docs locally & PySpark SQL tests
Author: Holden Karau <holden@us.ibm.com>
Closes#12938 from holdenk/SPARK-15162-SPARK-15164-update-some-pydocs.
#### What changes were proposed in this pull request?
This PR is to use the latest `SparkSession` to replace the existing `SQLContext` in `MLlib`. `SQLContext` is removed from `MLlib`.
Also fix a test case issue in `BroadcastJoinSuite`.
BTW, `SQLContext` is not being used in the `MLlib` test suites.
#### How was this patch tested?
Existing test cases.
Author: gatorsmile <gatorsmile@gmail.com>
Author: xiaoli <lixiao1983@gmail.com>
Author: Xiao Li <xiaoli@Xiaos-MacBook-Pro.local>
Closes#13380 from gatorsmile/sqlContextML.
## What changes were proposed in this pull request?
Deprecate `labelCol`, which is not used by ChiSqSelectorModel.
Author: Xiangrui Meng <meng@databricks.com>
Closes#13823 from mengxr/deprecate-setLabelCol-in-ChiSqSelectorModel.
## What changes were proposed in this pull request?
We forgot the getter of `dropLast` in `OneHotEncoder`
## How was this patch tested?
unit test
Author: Xiangrui Meng <meng@databricks.com>
Closes#13821 from mengxr/SPARK-16118.
## What changes were proposed in this pull request?
LibSVMFileFormat implements data source for LIBSVM format. However, users do not really need to call its APIs to use it. So we should hide it in the public API docs. The main issue is that we still need to put the documentation and example code somewhere. The proposal it to have a dummy class to hold the documentation, as a workaround to https://issues.scala-lang.org/browse/SI-8124.
## How was this patch tested?
Manually checked the generated API doc and tested loading LIBSVM data.
Author: Xiangrui Meng <meng@databricks.com>
Closes#13819 from mengxr/SPARK-16117.
## What changes were proposed in this pull request?
The `checkpointInterval` is a non-expert param. This PR moves its setter to non-expert group.
Author: Xiangrui Meng <meng@databricks.com>
Closes#13813 from mengxr/checkpoint-non-expert.
## What changes were proposed in this pull request?
This PR is a subset of #13023 by yanboliang to make SparkR model param names and default values consistent with MLlib. I tried to avoid other changes from #13023 to keep this PR minimal. I will send a follow-up PR to improve the documentation.
Main changes:
* `spark.glm`: epsilon -> tol, maxit -> maxIter
* `spark.kmeans`: default k -> 2, default maxIter -> 20, default initMode -> "k-means||"
* `spark.naiveBayes`: laplace -> smoothing, default 1.0
## How was this patch tested?
Existing unit tests.
Author: Xiangrui Meng <meng@databricks.com>
Closes#13801 from mengxr/SPARK-15177.1.
This PR adds missing `Since` annotations to `ml.feature` package.
Closes#8505.
## How was this patch tested?
Existing tests.
Author: Nick Pentreath <nickp@za.ibm.com>
Closes#13641 from MLnick/add-since-annotations.
## What changes were proposed in this pull request?
Both VectorUDT and MatrixUDT are private APIs, because UserDefinedType itself is private in Spark. However, in order to let developers implement their own transformers and estimators, we should expose both types in a public API to simply the implementation of transformSchema, transform, etc. Otherwise, they need to get the data types using reflection.
## How was this patch tested?
Unit tests in Scala and Java.
Author: Xiangrui Meng <meng@databricks.com>
Closes#13789 from mengxr/SPARK-16074.
## What changes were proposed in this pull request?
This PR implements python wrappers for #13662 to convert old/new vector columns in a DataFrame.
## How was this patch tested?
doctest in Python
cc: yanboliang
Author: Xiangrui Meng <meng@databricks.com>
Closes#13731 from mengxr/SPARK-15946.
JIRA: [SPARK-16008](https://issues.apache.org/jira/browse/SPARK-16008)
## What changes were proposed in this pull request?
`LogisticAggregator` stores references to two arrays of dimension `numFeatures` which are serialized before the combine op, unnecessarily. This results in the shuffle write being ~3x (for multiclass logistic regression, this number will go up) larger than it should be (in MLlib, for instance, it is 3x smaller).
This patch modifies `LogisticAggregator.add` to accept the two arrays as method parameters which avoids the serialization.
## How was this patch tested?
I tested this locally and verified the serialization reduction.
Additionally, I ran some tests of a 4 node cluster (4x48 cores, 4x128 GB RAM). Data set size of 2M rows and 10k features showed >2x iteration speedup.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#13729 from sethah/lr_improvement.
## What changes were proposed in this pull request?
SPARK-15922 reports the following scenario throwing an exception due to the mismatched vector sizes. This PR handles the exceptional case, `cols < (offset + colsPerBlock)`.
**Before**
```scala
scala> import org.apache.spark.mllib.linalg.distributed._
scala> import org.apache.spark.mllib.linalg._
scala> val rows = IndexedRow(0L, new DenseVector(Array(1,2,3))) :: IndexedRow(1L, new DenseVector(Array(1,2,3))):: IndexedRow(2L, new DenseVector(Array(1,2,3))):: Nil
scala> val rdd = sc.parallelize(rows)
scala> val matrix = new IndexedRowMatrix(rdd, 3, 3)
scala> val bmat = matrix.toBlockMatrix
scala> val imat = bmat.toIndexedRowMatrix
scala> imat.rows.collect
... // java.lang.IllegalArgumentException: requirement failed: Vectors must be the same length!
```
**After**
```scala
...
scala> imat.rows.collect
res0: Array[org.apache.spark.mllib.linalg.distributed.IndexedRow] = Array(IndexedRow(0,[1.0,2.0,3.0]), IndexedRow(1,[1.0,2.0,3.0]), IndexedRow(2,[1.0,2.0,3.0]))
```
## How was this patch tested?
Pass the Jenkins tests (including the above case)
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#13643 from dongjoon-hyun/SPARK-15922.
## What changes were proposed in this pull request?
Interface method `FileFormat.prepareRead()` was added in #12088 to handle a special case in the LibSVM data source.
However, the semantics of this interface method isn't intuitive: it returns a modified version of the data source options map. Considering that the LibSVM case can be easily handled using schema metadata inside `inferSchema`, we can remove this interface method to keep the `FileFormat` interface clean.
## How was this patch tested?
Existing tests.
Author: Cheng Lian <lian@databricks.com>
Closes#13698 from liancheng/remove-prepare-read.
## What changes were proposed in this pull request?
This patch renames various Parquet support classes from CatalystAbc to ParquetAbc. This new naming makes more sense for two reasons:
1. These are not optimizer related (i.e. Catalyst) classes.
2. We are in the Spark code base, and as a result it'd be more clear to call out these are Parquet support classes, rather than some Spark classes.
## How was this patch tested?
Renamed test cases as well.
Author: Reynold Xin <rxin@databricks.com>
Closes#13696 from rxin/parquet-rename.
The PR changes outdated scaladocs for Gini and Entropy classes. Since PR #886 Spark supports multiclass classification, but the docs tell only about binary classification.
Author: Wojciech Jurczyk <wojciech.jurczyk@codilime.com>
Closes#11252 from wjur/wjur/docs_multiclass.
## What changes were proposed in this pull request?
This PR provides conversion utils between old/new vector columns in a DataFrame. So users can use it to migrate their datasets and pipelines manually. The methods are implemented under `MLUtils` and called `convertVectorColumnsToML` and `convertVectorColumnsFromML`. Both take a DataFrame and a list of vector columns to be converted. It is a no-op on vector columns that are already converted. A warning message is logged if actual conversion happens.
This is the first sub-task under SPARK-15944 to make it easier to migrate existing pipelines to Spark 2.0.
## How was this patch tested?
Unit tests in Scala and Java.
cc: yanboliang
Author: Xiangrui Meng <meng@databricks.com>
Closes#13662 from mengxr/SPARK-15945.
## What changes were proposed in this pull request?
Now we have PySpark picklers for new and old vector/matrix, individually. However, they are all implemented under `PythonMLlibAPI`. To separate spark.mllib from spark.ml, we should implement the picklers of new vector/matrix under `spark.ml.python` instead.
## How was this patch tested?
Existing tests.
Author: Liang-Chi Hsieh <simonh@tw.ibm.com>
Closes#13219 from viirya/pyspark-pickler-ml.
## What changes were proposed in this pull request?
Currently, `AFTAggregator` is not being merged correctly. For example, if there is any single empty partition in the data, this creates an `AFTAggregator` with zero total count which causes the exception below:
```
IllegalArgumentException: u'requirement failed: The number of instances should be greater than 0.0, but got 0.'
```
Please see [AFTSurvivalRegression.scala#L573-L575](6ecedf39b4/mllib/src/main/scala/org/apache/spark/ml/regression/AFTSurvivalRegression.scala (L573-L575)) as well.
Just to be clear, the python example `aft_survival_regression.py` seems using 5 rows. So, if there exist partitions more than 5, it throws the exception above since it contains empty partitions which results in an incorrectly merged `AFTAggregator`.
Executing `bin/spark-submit examples/src/main/python/ml/aft_survival_regression.py` on a machine with CPUs more than 5 is being failed because it creates tasks with some empty partitions with defualt configurations (AFAIK, it sets the parallelism level to the number of CPU cores).
## How was this patch tested?
An unit test in `AFTSurvivalRegressionSuite.scala` and manually tested by `bin/spark-submit examples/src/main/python/ml/aft_survival_regression.py`.
Author: hyukjinkwon <gurwls223@gmail.com>
Author: Hyukjin Kwon <gurwls223@gmail.com>
Closes#13619 from HyukjinKwon/SPARK-15892.
## What changes were proposed in this pull request?
Currently, we always split the files when it's bigger than maxSplitBytes, but Hadoop LineRecordReader does not respect the splits for compressed files correctly, we should have a API for FileFormat to check whether the file could be splitted or not.
This PR is based on #13442, closes#13442
## How was this patch tested?
add regression tests.
Author: Davies Liu <davies@databricks.com>
Closes#13531 from davies/fix_split.
## What changes were proposed in this pull request?
In scala, immutable.List.length is an expensive operation so we should
avoid using Seq.length == 0 or Seq.lenth > 0, and use Seq.isEmpty and Seq.nonEmpty instead.
## How was this patch tested?
existing tests
Author: wangyang <wangyang@haizhi.com>
Closes#13601 from yangw1234/isEmpty.
## What changes were proposed in this pull request?
Adding __str__ to RFormula and model that will show the set formula param and resolved formula. This is currently present in the Scala API, found missing in PySpark during Spark 2.0 coverage review.
## How was this patch tested?
run pyspark-ml tests locally
Author: Bryan Cutler <cutlerb@gmail.com>
Closes#13481 from BryanCutler/pyspark-ml-rformula_str-SPARK-15738.
## What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-15793
Word2vec in ML package should have maxSentenceLength method for feature parity.
## How was this patch tested?
Tested with Spark unit test.
Author: yinxusen <yinxusen@gmail.com>
Closes#13536 from yinxusen/SPARK-15793.
## What changes were proposed in this pull request?
When fitting ```LinearRegressionModel```(by "l-bfgs" solver) and ```LogisticRegressionModel``` w/o intercept on dataset with constant nonzero column, spark.ml produce same model as R glmnet but different from LIBSVM.
When fitting ```AFTSurvivalRegressionModel``` w/o intercept on dataset with constant nonzero column, spark.ml produce different model compared with R survival::survreg.
We should output a warning message and clarify in document for this condition.
## How was this patch tested?
Document change, no unit test.
cc mengxr
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12731 from yanboliang/spark-13590.
## What changes were proposed in this pull request?
Made DefaultParamsReadable, DefaultParamsWritable public. Also added relevant doc and annotations. Added UnaryTransformerExample to demonstrate use of UnaryTransformer and DefaultParamsReadable,Writable.
## How was this patch tested?
Wrote example making use of the now-public APIs. Compiled and ran locally
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#13461 from jkbradley/defaultparamswritable.
## What changes were proposed in this pull request?
`an -> a`
Use cmds like `find . -name '*.R' | xargs -i sh -c "grep -in ' an [^aeiou]' {} && echo {}"` to generate candidates, and review them one by one.
## How was this patch tested?
manual tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#13515 from zhengruifeng/an_a.
`PartitionStatistics` uses `foldLeft` and list concatenation (`++`) to flatten an iterator of lists, but this is extremely inefficient compared to simply doing `flatMap`/`flatten` because it performs many unnecessary object allocations. Simply replacing this `foldLeft` by a `flatMap` results in decent performance gains when constructing PartitionStatistics instances for tables with many columns.
This patch fixes this and also makes two similar changes in MLlib and streaming to try to fix all known occurrences of this pattern.
Author: Josh Rosen <joshrosen@databricks.com>
Closes#13491 from JoshRosen/foldleft-to-flatmap.
## What changes were proposed in this pull request?
1, remove comments `:: Experimental ::` for non-experimental API
2, add comments `:: Experimental ::` for experimental API
3, add comments `:: DeveloperApi ::` for developerApi API
## How was this patch tested?
manual tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#13514 from zhengruifeng/del_experimental.
## What changes were proposed in this pull request?
1, del precision,recall in `ml.MulticlassClassificationEvaluator`
2, update user guide for `mlllib.weightedFMeasure`
## How was this patch tested?
local build
Author: Ruifeng Zheng <ruifengz@foxmail.com>
Closes#13390 from zhengruifeng/clarify_f1.
## What changes were proposed in this pull request?
Our encoder framework has been evolved a lot, this PR tries to clean up the code to make it more readable and emphasise the concept that encoder should be used as a container of serde expressions.
1. move validation logic to analyzer instead of encoder
2. only have a `resolveAndBind` method in encoder instead of `resolve` and `bind`, as we don't have the encoder life cycle concept anymore.
3. `Dataset` don't need to keep a resolved encoder, as there is no such concept anymore. bound encoder is still needed to do serialization outside of query framework.
4. Using `BoundReference` to represent an unresolved field in deserializer expression is kind of weird, this PR adds a `GetColumnByOrdinal` for this purpose. (serializer expression still use `BoundReference`, we can replace it with `GetColumnByOrdinal` in follow-ups)
## How was this patch tested?
existing test
Author: Wenchen Fan <wenchen@databricks.com>
Author: Cheng Lian <lian@databricks.com>
Closes#13269 from cloud-fan/clean-encoder.
## What changes were proposed in this pull request?
andrewor14 noticed some OOM errors caused by "test big model load / save" in Word2VecSuite, e.g., https://amplab.cs.berkeley.edu/jenkins/view/Spark%20QA%20Test/job/spark-master-test-maven-hadoop-2.2/1168/consoleFull. It doesn't show up in the test result because it was OOMed.
This PR disables the test. I will leave the JIRA open for a proper fix
## How was this patch tested?
No new features.
Author: Xiangrui Meng <meng@databricks.com>
Closes#13478 from mengxr/SPARK-15740.
## What changes were proposed in this pull request?
ml.feature: update check schema to avoid confusion when user use MLlib.vector as input type
## How was this patch tested?
existing ut
Author: Yuhao Yang <yuhao.yang@intel.com>
Closes#13411 from hhbyyh/schemaCheck.
Clean up style for param setter methods in ALS to match standard style and the other setter in class (this is an artefact of one of my previous PRs that wasn't cleaned up).
## How was this patch tested?
Existing tests - no functionality change.
Author: Nick Pentreath <nickp@za.ibm.com>
Closes#13480 from MLnick/als-param-minor-cleanup.
## What changes were proposed in this pull request?
ML 2.0 QA: Scala APIs audit for ml.feature. Mainly include:
* Remove seed for ```QuantileDiscretizer```, since we use ```approxQuantile``` to produce bins and ```seed``` is useless.
* Scala API docs update.
* Sync Scala and Python API docs for these changes.
## How was this patch tested?
Exist tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#13410 from yanboliang/spark-15587.
## What changes were proposed in this pull request?
if sparkContext.set CheckpointDir to another Dir that is not default FileSystem, it will throw exception when removing CheckpointFile in MLlib.
So we should always get the FileSystem from Path to avoid wrong FS problem.
## How was this patch tested?
N/A
Author: Lianhui Wang <lianhuiwang09@gmail.com>
Closes#13408 from lianhuiwang/SPARK-15664.
## What changes were proposed in this pull request?
This PR changes function `SparkSession.builder.sparkContext(..)` from **private[sql]** into **private[spark]**, and uses it if applicable like the followings.
```
- val spark = SparkSession.builder().config(sc.getConf).getOrCreate()
+ val spark = SparkSession.builder().sparkContext(sc).getOrCreate()
```
## How was this patch tested?
Pass the existing Jenkins tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#13365 from dongjoon-hyun/SPARK-15618.
## What changes were proposed in this pull request?
This change resolves a number of build warnings that have accumulated, before 2.x. It does not address a large number of deprecation warnings, especially related to the Accumulator API. That will happen separately.
## How was this patch tested?
Jenkins
Author: Sean Owen <sowen@cloudera.com>
Closes#13377 from srowen/BuildWarnings.
## What changes were proposed in this pull request?
Fix the wrong bound of `k` in `PCA`
`require(k <= sources.first().size, ...` -> `require(k < sources.first().size`
BTW, remove unused import in `ml.ElementwiseProduct`
## How was this patch tested?
manual tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#13356 from zhengruifeng/fix_pca.
## What changes were proposed in this pull request?
We're using `asML` to convert the mllib vector/matrix to ml vector/matrix now. Using `as` is more correct given that this conversion actually shares the same underline data structure. As a result, in this PR, `toBreeze` will be changed to `asBreeze`. This is a private API, as a result, it will not affect any user's application.
## How was this patch tested?
unit tests
Author: DB Tsai <dbt@netflix.com>
Closes#13198 from dbtsai/minor.
## What changes were proposed in this pull request?
* Document ```WeightedLeastSquares```(normal equation) and ```IterativelyReweightedLeastSquares```.
* Copy ```L-BFGS``` documents from ```spark.mllib``` to ```spark.ml```.
Due to the session ```Optimization of linear methods``` is used for developers, I think we should provide the brief introduction of the optimization method, necessary references and how it implements in Spark. It's not necessary to paste all mathematical formula and derivation here. If developers/users want to learn more, they can track reference.
## How was this patch tested?
Document update, no tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#13262 from yanboliang/spark-15484.
## What changes were proposed in this pull request?
This PR replaces `spark.sql.sources.` strings with `CreateDataSourceTableUtils.*` constant variables.
## How was this patch tested?
Pass the existing Jenkins tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#13349 from dongjoon-hyun/SPARK-15584.
## What changes were proposed in this pull request?
This PR replaces all deprecated `SQLContext` occurrences with `SparkSession` in `ML/MLLib` module except the following two classes. These two classes use `SQLContext` in their function signatures.
- ReadWrite.scala
- TreeModels.scala
## How was this patch tested?
Pass the existing Jenkins tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#13352 from dongjoon-hyun/SPARK-15603.
## What changes were proposed in this pull request?
`a` -> `an`
I use regex to generate potential error lines:
`grep -in ' a [aeiou]' mllib/src/main/scala/org/apache/spark/ml/*/*scala`
and review them line by line.
## How was this patch tested?
local build
`lint-java` checking
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#13317 from zhengruifeng/a_an.
## What changes were proposed in this pull request?
This PR changes SQLContext/HiveContext's public constructor to use SparkSession.build.getOrCreate and removes isRootContext from SQLContext.
## How was this patch tested?
Existing tests.
Author: Yin Huai <yhuai@databricks.com>
Closes#13310 from yhuai/SPARK-15532.
## What changes were proposed in this pull request?
Several classes and methods have been deprecated and are creating lots of build warnings in branch-2.0. This issue is to identify and fix those items:
* WithSGD classes: Change to make class not deprecated, object deprecated, and public class constructor deprecated. Any public use will require a deprecated API. We need to keep a non-deprecated private API since we cannot eliminate certain uses: Python API, streaming algs, and examples.
* Use in PythonMLlibAPI: Change to using private constructors
* Streaming algs: No warnings after we un-deprecate the classes
* Examples: Deprecate or change ones which use deprecated APIs
* MulticlassMetrics fields (precision, etc.)
* LinearRegressionSummary.model field
## How was this patch tested?
Existing tests. Checked for warnings manually.
Author: Sean Owen <sowen@cloudera.com>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#13314 from jkbradley/warning-cleanups.
## What changes were proposed in this pull request?
See https://issues.apache.org/jira/browse/SPARK-15523
This PR replaces PR #13293. It's isolated to a new branch, and contains some more squashed changes.
## How was this patch tested?
1. Executed `mvn clean package` in `mllib` directory
2. Executed `dev/test-dependencies.sh --replace-manifest` in the root directory.
Author: Villu Ruusmann <villu.ruusmann@gmail.com>
Closes#13297 from vruusmann/update-jpmml.
## What changes were proposed in this pull request?
This patch renames various DefaultSources to make their names more self-describing. The choice of "DefaultSource" was from the days when we did not have a good way to specify short names.
They are now named:
- LibSVMFileFormat
- CSVFileFormat
- JdbcRelationProvider
- JsonFileFormat
- ParquetFileFormat
- TextFileFormat
Backward compatibility is maintained through aliasing.
## How was this patch tested?
Updated relevant test cases too.
Author: Reynold Xin <rxin@databricks.com>
Closes#13311 from rxin/SPARK-15543.
## What changes were proposed in this pull request?
Add a warning log for the case that `numIterations * miniBatchFraction <1.0` during gradient descent. If the product of those two numbers is less than `1.0`, then not all training examples will be used during optimization. To put this concretely, suppose that `numExamples = 100`, `miniBatchFraction = 0.2` and `numIterations = 3`. Then, 3 iterations will occur each sampling approximately 6 examples each. In the best case, each of the 6 examples are unique; hence 18/100 examples are used.
This may be counter-intuitive to most users and led to the issue during the development of another Spark ML model: https://github.com/zhengruifeng/spark-libFM/issues/11. If a user actually does not require the training data set, it would be easier and more intuitive to use `RDD.sample`.
## How was this patch tested?
`build/mvn -DskipTests clean package` build succeeds
Author: Gio Borje <gborje@linkedin.com>
Closes#13265 from Hydrotoast/master.
Remove "Default: MEMORY_AND_DISK" from `Param` doc field in ALS storage level params. This fixes up the output of `explainParam(s)` so that default values are not displayed twice.
We can revisit in the case that [SPARK-15130](https://issues.apache.org/jira/browse/SPARK-15130) moves ahead with adding defaults in some way to PySpark param doc fields.
Tests N/A.
Author: Nick Pentreath <nickp@za.ibm.com>
Closes#13277 from MLnick/SPARK-15500-als-remove-default-storage-param.
fixed typos for source code for components [mllib] [streaming] and [SQL]
None and obvious.
Author: lfzCarlosC <lfz.carlos@gmail.com>
Closes#13298 from lfzCarlosC/master.
This PR adds the `relativeError` param to PySpark's `QuantileDiscretizer` to match Scala.
Also cleaned up a duplication of `numBuckets` where the param is both a class and instance attribute (I removed the instance attr to match the style of params throughout `ml`).
Finally, cleaned up the docs for `QuantileDiscretizer` to reflect that it now uses `approxQuantile`.
## How was this patch tested?
A little doctest and built API docs locally to check HTML doc generation.
Author: Nick Pentreath <nickp@za.ibm.com>
Closes#13228 from MLnick/SPARK-15442-py-relerror-param.
## What changes were proposed in this pull request?
* ```GeneralizedLinearRegression``` API docs enhancement.
* The default value of ```GeneralizedLinearRegression``` ```linkPredictionCol``` is not set rather than empty. This will consistent with other similar params such as ```weightCol```
* Make some methods more private.
* Fix a minor bug of LinearRegression.
* Fix some other issues.
## How was this patch tested?
Existing tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#13129 from yanboliang/spark-15339.
## What changes were proposed in this pull request?
Currently SparkSession.Builder use SQLContext.getOrCreate. It should probably the the other way around, i.e. all the core logic goes in SparkSession, and SQLContext just calls that. This patch does that.
This patch also makes sure config options specified in the builder are propagated to the existing (and of course the new) SparkSession.
## How was this patch tested?
Updated tests to reflect the change, and also introduced a new SparkSessionBuilderSuite that should cover all the branches.
Author: Reynold Xin <rxin@databricks.com>
Closes#13200 from rxin/SPARK-15075.
## What changes were proposed in this pull request?
Refactor All Java Tests that use SparkSession, to extend SharedSparkSesion
## How was this patch tested?
Existing Tests
Author: Sandeep Singh <sandeep@techaddict.me>
Closes#13101 from techaddict/SPARK-15296.
## What changes were proposed in this pull request?
```ml.evaluation``` Scala and Python API sync.
## How was this patch tested?
Only API docs change, no new tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#13195 from yanboliang/evaluation-doc.
## What changes were proposed in this pull request?
Currently in ```model.write```, we don't save ```summary```(if applicable). We should add documentation to clarify it.
We fixed the incorrect link ```[[MLWriter]]``` to ```[[org.apache.spark.ml.util.MLWriter]]``` BTW.
## How was this patch tested?
Documentation update, no unit test.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#13131 from yanboliang/spark-15341.
## What changes were proposed in this pull request?
Open up APIs for converting between new, old linear algebra types (in spark.mllib.linalg):
`Sparse`/`Dense` X `Vector`/`Matrices` `.asML` and `.fromML`
## How was this patch tested?
Existing Tests
Author: Sandeep Singh <sandeep@techaddict.me>
Closes#13202 from techaddict/SPARK-15414.
## What changes were proposed in this pull request?
Audit Scala API for ml.clustering.
Fix some wrong API documentations and update outdated one.
## How was this patch tested?
Existing unit tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#13148 from yanboliang/spark-15361.
## What changes were proposed in this pull request?
Add since to ml.stat.MultivariateOnlineSummarizer.scala
## How was this patch tested?
unit tests
Author: DB Tsai <dbt@netflix.com>
Closes#13197 from dbtsai/cleanup.
## What changes were proposed in this pull request?
Audit Scala API for classification, almost all issues were related ```MultilayerPerceptronClassifier``` in this section.
* Fix one wrong param getter function: ```getOptimizer``` -> ```getSolver```
* Add missing setter function for ```solver``` and ```stepSize```.
* Make ```GD``` solver take effect.
* Update docs, annotations and fix other minor issues.
## How was this patch tested?
Existing unit tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#13076 from yanboliang/spark-15292.
## What changes were proposed in this pull request?
[SPARK-14646](https://issues.apache.org/jira/browse/SPARK-14646) makes ```KMeansModel``` store the cluster centers one per row. ```KMeansModel.load()``` method needs to be updated in order to load models saved with Spark 1.6.
## How was this patch tested?
Since ```save/load``` is ```Experimental``` for 1.6, I think offline test for backwards compatibility is enough.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#13149 from yanboliang/spark-15362.
## What changes were proposed in this pull request?
I reviewed Scala and Python APIs for ml.feature and corrected discrepancies.
## How was this patch tested?
Built docs locally, ran style checks
Author: Bryan Cutler <cutlerb@gmail.com>
Closes#13159 from BryanCutler/ml.feature-api-sync.
## What changes were proposed in this pull request?
This PR adds null check in `SparkSession.createDataFrame`, so that we can make sure the passed in rows matches the given schema.
## How was this patch tested?
new tests in `DatasetSuite`
Author: Wenchen Fan <wenchen@databricks.com>
Closes#13008 from cloud-fan/row-encoder.
This PR adds schema validation to `ml`'s ALS and ALSModel. Currently, no schema validation was performed as `transformSchema` was never called in `ALS.fit` or `ALSModel.transform`. Furthermore, due to no schema validation, if users passed in Long (or Float etc) ids, they would be silently cast to Int with no warning or error thrown.
With this PR, ALS now supports all numeric types for `user`, `item`, and `rating` columns. The rating column is cast to `Float` and the user and item cols are cast to `Int` (as is the case currently) - however for user/item, the cast throws an error if the value is outside integer range. Behavior for rating col is unchanged (as it is not an issue).
## How was this patch tested?
New test cases in `ALSSuite`.
Author: Nick Pentreath <nickp@za.ibm.com>
Closes#12762 from MLnick/SPARK-14891-als-validate-schema.
mateiz srowen
I state that the contribution is my original work and that I license the work to the project under the project's open source license
There's some format problems with my last PR, with HyukjinKwon 's help I read the guidance, re-check my code and PR, then run the tests, finally re-submit the PR request here.
The related JIRA issue though marked as resolved, this change may relate to it I think.
## Proposed Change
After picking each new initial centers, it's unnecessary to compute the distances between all the points and the old ones.
Instead this change keeps the distance between all the points and their closest centers, and compare to the distance of them with the new center then update them.
## Test result
One can find an easy test way in (https://issues.apache.org/jira/browse/SPARK-6706)
I test the KMeans++ method for a small dataset with 16k points, and the whole KMeans|| with a large one with 240k points.
The data has 4096 features and I tunes K from 100 to 500.
The test environment was on my 4 machine cluster, I also tested a 3M points data on a larger cluster with 25 machines and got similar results, which I would not draw the detail curve. The result of the first two exps are shown below
### Local KMeans++ test:
Dataset:4m_ini_center
Data_size:16234
Dimension:4096
Lloyd's Iteration = 10
The y-axis is time in sec, the x-axis is tuning the K.
### On a larger dataset
An improve show in the graph but not commit in this file: In this experiment I also have an improvement for calculation in normalization data (the distance is convert to the cosine distance). As if the data is normalized into (0,1), one improvement in the original vesion for util.MLUtils.fastSauaredDistance would have no effect (the precisionBound 2.0 * EPSILON * sumSquaredNorm / (normDiff * normDiff + EPSILON) will never less then precision in this case). Therefore I design an early terminal method when comparing two distance (used for findClosest). But I don't include this improve in this file, you may only refer to the curves without "normalize" for comparing the results.
Dataset:4k24
Data_size:243960
Dimension:4096
Normlize Enlarge Initialize Lloyd's_Iteration
NO 1 3 5
YES 10000 3 5
Notice: the normlized data is enlarged to ensure precision
The cost time: x-for value of K, y-for time in sec
SE for unnormalized data between two version, to ensure the correctness
Here is the SE between normalized data just for reference, it's also correct.
Author: DLucky <mouendless@gmail.com>
Closes#13133 from mouendless/patch-2.
## What changes were proposed in this pull request?
I use Intellj-IDEA to search usage of deprecate SparkContext.accumulator in the whole spark project, and update the code.(except those test code for accumulator method itself)
## How was this patch tested?
Exisiting unit tests
Author: WeichenXu <WeichenXu123@outlook.com>
Closes#13112 from WeichenXu123/update_accuV2_in_mllib.
## What changes were proposed in this pull request?
Update the unit test code, examples, and documents to remove calls to deprecated method `dataset.registerTempTable`.
## How was this patch tested?
This PR only changes the unit test code, examples, and comments. It should be safe.
This is a follow up of PR https://github.com/apache/spark/pull/12945 which was merged.
Author: Sean Zhong <seanzhong@databricks.com>
Closes#13098 from clockfly/spark-15171-remove-deprecation.
## What changes were proposed in this pull request?
Once SPARK-14487 and SPARK-14549 are merged, we will migrate to use the new vector and matrix type in the new ml pipeline based apis.
## How was this patch tested?
Unit tests
Author: DB Tsai <dbt@netflix.com>
Author: Liang-Chi Hsieh <simonh@tw.ibm.com>
Author: Xiangrui Meng <meng@databricks.com>
Closes#12627 from dbtsai/SPARK-14615-NewML.
## What changes were proposed in this pull request?
According to the recent change, this PR replaces all the remaining `sqlContext` usage with `spark` in ScalaDoc/JavaDoc (.scala/.java files) except `SQLContext.scala`, `SparkPlan.scala', and `DatasetHolder.scala`.
## How was this patch tested?
Manual.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#13125 from dongjoon-hyun/minor_doc_sparksession.
## What changes were proposed in this pull request?
(See https://github.com/apache/spark/pull/12416 where most of this was already reviewed and committed; this is just the module structure and move part. This change does not move the annotations into test scope, which was the apparently problem last time.)
Rename `spark-test-tags` -> `spark-tags`; move common annotations like `Since` to `spark-tags`
## How was this patch tested?
Jenkins tests.
Author: Sean Owen <sowen@cloudera.com>
Closes#13074 from srowen/SPARK-15290.
## What changes were proposed in this pull request?
1,Rename matrix args in BreezeUtil to upper to match the doc
2,Fix several typos in ML and SQL
## How was this patch tested?
manual tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#13078 from zhengruifeng/fix_ann.
## What changes were proposed in this pull request?
(Please fill in changes proposed in this fix)
Throw better exception when numClasses is empty and empty.max is thrown.
## How was this patch tested?
(Please explain how this patch was tested. E.g. unit tests, integration tests, manual tests)
Add a new unit test, which calls histogram with empty numClasses.
Author: wm624@hotmail.com <wm624@hotmail.com>
Closes#12969 from wangmiao1981/logisticR.
## What changes were proposed in this pull request?
Currently, Parquet, JSON and CSV data sources have a class for thier options, (`ParquetOptions`, `JSONOptions` and `CSVOptions`).
It is convenient to manage options for sources to gather options into a class. Currently, `JDBC`, `Text`, `libsvm` and `ORC` datasources do not have this class. This might be nicer if these options are in a unified format so that options can be added and
This PR refactors the options in Spark internal data sources adding new classes, `OrcOptions`, `TextOptions`, `JDBCOptions` and `LibSVMOptions`.
Also, this PR change the default compression codec for ORC from `NONE` to `SNAPPY`.
## How was this patch tested?
Existing tests should cover this for refactoring and unittests in `OrcHadoopFsRelationSuite` for changing the default compression codec for ORC.
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#13048 from HyukjinKwon/SPARK-15267.
## What changes were proposed in this pull request?
(Please fill in changes proposed in this fix)
Add accuracy to MulticlassMetrics class and add corresponding code in MulticlassClassificationEvaluator.
## How was this patch tested?
(Please explain how this patch was tested. E.g. unit tests, integration tests, manual tests)
Scala Unit tests in ml.evaluation
Author: wm624@hotmail.com <wm624@hotmail.com>
Closes#12882 from wangmiao1981/accuracy.
## What changes were proposed in this pull request?
Made ChiSqSelector and RFormula accept all numeric types for label
## How was this patch tested?
Unit tests
Author: BenFradet <benjamin.fradet@gmail.com>
Closes#12467 from BenFradet/SPARK-13961.
## What changes were proposed in this pull request?
This patch adds a python API for generalized linear regression summaries (training and test). This helps provide feature parity for Python GLMs.
## How was this patch tested?
Added a unit test to `pyspark.ml.tests`
Author: sethah <seth.hendrickson16@gmail.com>
Closes#12961 from sethah/GLR_summary.
## What changes were proposed in this pull request?
Deprecates registerTempTable and add dataset.createTempView, dataset.createOrReplaceTempView.
## How was this patch tested?
Unit tests.
Author: Sean Zhong <seanzhong@databricks.com>
Closes#12945 from clockfly/spark-15171.
## What changes were proposed in this pull request?
We have a private `UDTRegistration` API to register user defined type. Currently `JavaTypeInference` can't work with it. So `SparkSession.createDataFrame` from a bean class will not correctly infer the schema of the bean class.
## How was this patch tested?
`VectorUDTSuite`.
Author: Liang-Chi Hsieh <simonh@tw.ibm.com>
Closes#13046 from viirya/fix-udt-registry-javatypeinference.
## What changes were proposed in this pull request?
Use SparkSession instead of SQLContext in Scala/Java TestSuites
as this PR already very big working Python TestSuites in a diff PR.
## How was this patch tested?
Existing tests
Author: Sandeep Singh <sandeep@techaddict.me>
Closes#12907 from techaddict/SPARK-15037.
## What changes were proposed in this pull request?
Explicitly tell user initial coefficients is ignored if its size doesn't match expected size in LogisticRegression
## How was this patch tested?
local build
Author: dding3 <dingding@dingding-ubuntu.sh.intel.com>
Closes#12948 from dding3/master.
## What changes were proposed in this pull request?
jira: https://issues.apache.org/jira/browse/SPARK-14814
fix a java compatibility function in mllib DecisionTreeModel. As synced in jira, other compatibility issues don't need fixes.
## How was this patch tested?
existing ut
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#12971 from hhbyyh/javacompatibility.
## What changes were proposed in this pull request?
We need to use `requiredSchema` in `LibSVMRelation` to project the fetch required columns when loading data from this data source. Otherwise, when users try to select `features` column, it will cause failure.
## How was this patch tested?
`LibSVMRelationSuite`.
Author: Liang-Chi Hsieh <simonh@tw.ibm.com>
Closes#12986 from viirya/fix-libsvmrelation.
## What changes were proposed in this pull request?
This PR continues the work from #11871 with the following changes:
* load English stopwords as default
* covert stopwords to list in Python
* update some tests and doc
## How was this patch tested?
Unit tests.
Closes#11871
cc: burakkose srowen
Author: Burak Köse <burakks41@gmail.com>
Author: Xiangrui Meng <meng@databricks.com>
Author: Burak KOSE <burakks41@gmail.com>
Closes#12843 from mengxr/SPARK-14050.
## What changes were proposed in this pull request?
Minor doc and code style fixes
## How was this patch tested?
local build
Author: Jacek Laskowski <jacek@japila.pl>
Closes#12928 from jaceklaskowski/SPARK-15152.
## What changes were proposed in this pull request?
Copy the package documentation from Scala/Java to Python for ML package and remove beta tags. Not super sure if we want to keep the BETA tag but since we are making it the default it seems like probably the time to remove it (happy to put it back in if we want to keep it BETA).
## How was this patch tested?
Python documentation built locally as HTML and text and verified output.
Author: Holden Karau <holden@us.ibm.com>
Closes#12883 from holdenk/SPARK-15106-add-pyspark-package-doc-for-ml.
## What changes were proposed in this pull request?
Introduction of setFeaturesCol and setPredictionCol methods to KMeansModel in ML library.
## How was this patch tested?
By running KMeansSuite.
Author: Dominik Jastrzębski <dominik.jastrzebski@codilime.com>
Closes#12609 from dominik-jastrzebski/master.
## What changes were proposed in this pull request?
Currently, various `FileFormat` data sources share approximately the same code for partition value appending. This PR tries to eliminate this duplication.
A new method `buildReaderWithPartitionValues()` is added to `FileFormat` with a default implementation that appends partition values to `InternalRow`s produced by the reader function returned by `buildReader()`.
Special data sources like Parquet, which implements partition value appending inside `buildReader()` because of the vectorized reader, and the Text data source, which doesn't support partitioning, override `buildReaderWithPartitionValues()` and simply delegate to `buildReader()`.
This PR brings two benefits:
1. Apparently, it de-duplicates partition value appending logic
2. Now the reader function returned by `buildReader()` is only required to produce `InternalRow`s rather than `UnsafeRow`s if the data source doesn't override `buildReaderWithPartitionValues()`.
Because the safe-to-unsafe conversion is also performed while appending partition values. This makes 3rd-party data sources (e.g. spark-avro) easier to implement since they no longer need to access private APIs involving `UnsafeRow`.
## How was this patch tested?
Existing tests should do the work.
Author: Cheng Lian <lian@databricks.com>
Closes#12866 from liancheng/spark-14237-simplify-partition-values-appending.
## What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-14973
Add seed support when saving/loading of CrossValidator and TrainValidationSplit.
## How was this patch tested?
Spark unit test.
Author: yinxusen <yinxusen@gmail.com>
Closes#12825 from yinxusen/SPARK-14973.
## What changes were proposed in this pull request?
When ALS is run with a checkpoint interval, during the checkpoint materialize the current state and cleanup the previous shuffles (non-blocking).
## How was this patch tested?
Existing ALS unit tests, new ALS checkpoint cleanup unit tests added & shuffle files checked after ALS w/checkpointing run.
Author: Holden Karau <holden@us.ibm.com>
Author: Holden Karau <holden@pigscanfly.ca>
Closes#11919 from holdenk/SPARK-6717-clear-shuffle-files-after-checkpointing-in-ALS.
## What changes were proposed in this pull request?
This PR is an update for [https://github.com/apache/spark/pull/12738] which:
* Adds a generic unit test for JavaParams wrappers in pyspark.ml for checking default Param values vs. the defaults in the Scala side
* Various fixes for bugs found
* This includes changing classes taking weightCol to treat unset and empty String Param values the same way.
Defaults changed:
* Scala
* LogisticRegression: weightCol defaults to not set (instead of empty string)
* StringIndexer: labels default to not set (instead of empty array)
* GeneralizedLinearRegression:
* maxIter always defaults to 25 (simpler than defaulting to 25 for a particular solver)
* weightCol defaults to not set (instead of empty string)
* LinearRegression: weightCol defaults to not set (instead of empty string)
* Python
* MultilayerPerceptron: layers default to not set (instead of [1,1])
* ChiSqSelector: numTopFeatures defaults to 50 (instead of not set)
## How was this patch tested?
Generic unit test. Manually tested that unit test by changing defaults and verifying that broke the test.
Author: Joseph K. Bradley <joseph@databricks.com>
Author: yinxusen <yinxusen@gmail.com>
Closes#12816 from jkbradley/yinxusen-SPARK-14931.
## What changes were proposed in this pull request?
* ```RFormula``` supports empty response variable like ```~ x + y```.
* Support formula in ```spark.kmeans``` in SparkR.
* Fix some outdated docs for SparkR.
## How was this patch tested?
Unit tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12813 from yanboliang/spark-15030.
#### What changes were proposed in this pull request?
This PR removes three methods the were deprecated in 1.6.0:
- `PortableDataStream.close()`
- `LinearRegression.weights`
- `LogisticRegression.weights`
The rationale for doing this is that the impact is small and that Spark 2.0 is a major release.
#### How was this patch tested?
Compilation succeded.
Author: Herman van Hovell <hvanhovell@questtec.nl>
Closes#12732 from hvanhovell/SPARK-14952.
## What changes were proposed in this pull request?
This PR moves Vector.toJson/fromJson to ml.linalg.VectorEncoder under mllib/ to keep mllib-local's dependency minimal. The json encoding is used by Params. So we still need this feature in SPARK-14615, where we will switch to ml.linalg in spark.ml APIs.
## How was this patch tested?
Copied existing unit tests over.
cc; dbtsai
Author: Xiangrui Meng <meng@databricks.com>
Closes#12802 from mengxr/SPARK-14653.
## What changes were proposed in this pull request?
This PR fixes the bug that generates infinite distances between word vectors. For example,
Before this PR, we have
```
val synonyms = model.findSynonyms("who", 40)
```
will give the following results:
```
to Infinity
and Infinity
that Infinity
with Infinity
```
With this PR, the distance between words is a value between 0 and 1, as follows:
```
scala> model.findSynonyms("who", 10)
res0: Array[(String, Double)] = Array((Harvard-educated,0.5253688097000122), (ex-SAS,0.5213794708251953), (McMutrie,0.5187736749649048), (fellow,0.5166833400726318), (businessman,0.5145374536514282), (American-born,0.5127736330032349), (British-born,0.5062344074249268), (gray-bearded,0.5047978162765503), (American-educated,0.5035858750343323), (mentored,0.49849334359169006))
scala> model.findSynonyms("king", 10)
res1: Array[(String, Double)] = Array((queen,0.6787897944450378), (prince,0.6786158084869385), (monarch,0.659771203994751), (emperor,0.6490438580513), (goddess,0.643266499042511), (dynasty,0.635733425617218), (sultan,0.6166239380836487), (pharaoh,0.6150713562965393), (birthplace,0.6143025159835815), (empress,0.6109727025032043))
scala> model.findSynonyms("queen", 10)
res2: Array[(String, Double)] = Array((princess,0.7670737504959106), (godmother,0.6982434988021851), (raven-haired,0.6877717971801758), (swan,0.684934139251709), (hunky,0.6816608309745789), (Titania,0.6808111071586609), (heroine,0.6794036030769348), (king,0.6787897944450378), (diva,0.67848801612854), (lip-synching,0.6731793284416199))
```
### There are two places changed in this PR:
- Normalize the word vector to avoid overflow when calculating inner product between word vectors. This also simplifies the distance calculation, since the word vectors only need to be normalized once.
- Scale the learning rate by number of iteration, to be consistent with Google Word2Vec implementation
## How was this patch tested?
Use word2vec to train text corpus, and run model.findSynonyms() to get the distances between word vectors.
Author: Junyang <fly.shenjy@gmail.com>
Author: flyskyfly <fly.shenjy@gmail.com>
Closes#11812 from flyjy/TVec.
## What changes were proposed in this pull request?
As discussed in #12660, this PR renames
* intermediateRDDStorageLevel -> intermediateStorageLevel
* finalRDDStorageLevel -> finalStorageLevel
The argument name in `ALS.train` will be addressed in SPARK-15027.
## How was this patch tested?
Existing unit tests.
Author: Xiangrui Meng <meng@databricks.com>
Closes#12803 from mengxr/SPARK-14412.
## What changes were proposed in this pull request?
Fix for part of SPARK-14533: trivial simplification and more accurate computation of column means. See also https://github.com/apache/spark/pull/12299 which contained a complete fix that was very slow. This PR does _not_ resolve SPARK-14533 entirely.
## How was this patch tested?
Existing tests.
Author: Sean Owen <sowen@cloudera.com>
Closes#12779 from srowen/SPARK-14533.2.
## What changes were proposed in this pull request?
This PR uses `UnsafeArrayData.fromPrimitiveArray` to implement `ml.VectorUDT/MatrixUDT` to avoid boxing/unboxing.
## How was this patch tested?
Exiting unit tests.
cc: cloud-fan
Author: Xiangrui Meng <meng@databricks.com>
Closes#12805 from mengxr/SPARK-14850.
## What changes were proposed in this pull request?
This PR adds `fromPrimitiveArray` and `toPrimitiveArray` in `UnsafeArrayData`, so that we can do the conversion much faster in VectorUDT/MatrixUDT.
## How was this patch tested?
existing tests and new test suite `UnsafeArraySuite`
Author: Wenchen Fan <wenchen@databricks.com>
Closes#12640 from cloud-fan/ml.
`mllib` `ALS` supports `setIntermediateRDDStorageLevel` and `setFinalRDDStorageLevel`. This PR adds these as Params in `ml` `ALS`. They are put in group **expertParam** since few users will need them.
## How was this patch tested?
New test cases in `ALSSuite` and `tests.py`.
cc yanboliang jkbradley sethah rishabhbhardwaj
Author: Nick Pentreath <nickp@za.ibm.com>
Closes#12660 from MLnick/SPARK-14412-als-storage-params.
## What changes were proposed in this pull request?
Modified Kmeans to store cluster centers with one per row
## How was this patch tested?
Existing tests
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12792 from jkbradley/kmeans-save-fix.
## What changes were proposed in this pull request?
Added Instrumentation logging to DecisionTree{Classifier,Regressor} and RandomForest{Classifier,Regressor}
## How was this patch tested?
No tests involved since it's logging related.
Author: BenFradet <benjamin.fradet@gmail.com>
Closes#12536 from BenFradet/SPARK-14570.
## What changes were proposed in this pull request?
pyspark.ml API for LDA
* LDA, LDAModel, LocalLDAModel, DistributedLDAModel
* includes persistence
This replaces [https://github.com/apache/spark/pull/10242]
## How was this patch tested?
* doc test for LDA, including Param setters
* unit test for persistence
Author: Joseph K. Bradley <joseph@databricks.com>
Author: Jeff Zhang <zjffdu@apache.org>
Closes#12723 from jkbradley/zjffdu-SPARK-11940.
## What changes were proposed in this pull request?
Deprecated model field in LinearRegressionSummary
Removed unnecessary Since annotations
## How was this patch tested?
Existing tests
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12763 from jkbradley/lr-summary-api.
SparkR ```glm``` and ```kmeans``` model persistence.
Unit tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Author: Gayathri Murali <gayathri.m.softie@gmail.com>
Closes#12778 from yanboliang/spark-14311.
Closes#12680Closes#12683
## What changes were proposed in this pull request?
Add log instrumentation for parameters:
rank, numUserBlocks, numItemBlocks, implicitPrefs, alpha,
userCol, itemCol, ratingCol, predictionCol, maxIter,
regParam, nonnegative, checkpointInterval, seed
Add log instrumentation for numUserFeatures and numItemFeatures
## How was this patch tested?
Manual test: Set breakpoint in intellij and run def testALS(). Single step debugging and check the log method is called.
Author: wm624@hotmail.com <wm624@hotmail.com>
Closes#12560 from wangmiao1981/log.
## What changes were proposed in this pull request?
This PR removes duplicate implementation of compute in LogisticGradient class
## How was this patch tested?
unit tests
Author: dding3 <dingding@dingding-ubuntu.sh.intel.com>
Closes#12747 from dding3/master.
## What changes were proposed in this pull request?
Handle case where number of predictions is less than label set, k in nDCG computation
## How was this patch tested?
New unit test; existing tests
Author: Sean Owen <sowen@cloudera.com>
Closes#12756 from srowen/SPARK-14886.
## What changes were proposed in this pull request?
According to the [SPARK-14829](https://issues.apache.org/jira/browse/SPARK-14829), deprecate API of LogisticRegression and LinearRegression using SGD
## How was this patch tested?
manual tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#12596 from zhengruifeng/deprecate_sgd.
## What changes were proposed in this pull request?
Updated Classifier, DecisionTreeClassifier, RandomForestClassifier, GBTClassifier to not require input column metadata.
* They first check for metadata.
* If numClasses is not specified in metadata, they identify the largest label value (up to a limit).
This functionality is implemented in a new Classifier.getNumClasses method.
Also
* Updated Classifier.extractLabeledPoints to (a) check label values and (b) include a second version which takes a numClasses value for validity checking.
## How was this patch tested?
* Unit tests in ClassifierSuite for helper methods
* Unit tests for DecisionTreeClassifier, RandomForestClassifier, GBTClassifier with toy datasets lacking label metadata
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12663 from jkbradley/trees-no-metadata.
## What changes were proposed in this pull request?
This PR adds `since` tag into the matrix and vector classes in spark-mllib-local.
## How was this patch tested?
Scala-style checks passed.
Author: Pravin Gadakh <prgadakh@in.ibm.com>
Closes#12416 from pravingadakh/SPARK-14613.
## What changes were proposed in this pull request?
jira: https://issues.apache.org/jira/browse/SPARK-14916
FreqItemset as the result of FPGrowth should have a more friendly toString(), to help users and developers.
sample:
{a, b}: 5
{x, y, z}: 4
## How was this patch tested?
existing unit tests.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#12698 from hhbyyh/freqtos.
## What changes were proposed in this pull request?
This splits GeneralizedLinearRegressionSummary into 2 summary types:
* GeneralizedLinearRegressionSummary, which does not store info from fitting (diagInvAtWA)
* GeneralizedLinearRegressionTrainingSummary, which is a subclass of GeneralizedLinearRegressionSummary and stores info from fitting
This also add a method evaluate() which can produce a GeneralizedLinearRegressionSummary on a new dataset.
The summary no longer provides the model itself as a public val.
Also:
* Fixes bug where GeneralizedLinearRegressionTrainingSummary was created with model, not summaryModel.
* Adds hasSummary method.
* Renames findSummaryModelAndPredictionCol -> getSummaryModel and simplifies that method.
* In summary, extract values from model immediately in case user later changes those (e.g., predictionCol).
* Pardon the style fixes; that is IntelliJ being obnoxious.
## How was this patch tested?
Existing unit tests + updated test for evaluate and hasSummary
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12624 from jkbradley/model-summary-api.
## What changes were proposed in this pull request?
Currently we use `SQLUserDefinedType` annotation to register UDTs for user classes. However, by doing this, we add Spark dependency to user classes.
For some user classes, it is unnecessary to add such dependency that will increase deployment difficulty.
We should provide alternative approach to register UDTs for user classes without `SQLUserDefinedType` annotation.
## How was this patch tested?
`UserDefinedTypeSuite`
Author: Liang-Chi Hsieh <simonh@tw.ibm.com>
Closes#12259 from viirya/improve-sql-usertype.
## What changes were proposed in this pull request?
Pipeline.setStages failed for some code examples which worked in 1.5 but fail in 1.6. This tends to occur when using a mix of transformers from ml.feature. It is because Java Arrays are non-covariant and the addition of MLWritable to some transformers means the stages0/1 arrays above are not of type Array[PipelineStage]. This PR modifies the following to accept subclasses of PipelineStage:
* Pipeline.setStages()
* Params.w()
## How was this patch tested?
Unit test which fails to compile before this fix.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12430 from jkbradley/pipeline-setstages.
## What changes were proposed in this pull request?
Since [SPARK-10574](https://issues.apache.org/jira/browse/SPARK-10574) breaks behavior of ```HashingTF```, we should try to enforce good practice by removing the "native" hashAlgorithm option in spark.ml and pyspark.ml. We can leave spark.mllib and pyspark.mllib alone.
## How was this patch tested?
Unit tests.
cc jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12702 from yanboliang/spark-14899.
This PR adds the remaining group of methods to PySpark's distributed linear algebra classes as follows:
* `RowMatrix` <sup>**[1]**</sup>
1. `computeGramianMatrix`
2. `computeCovariance`
3. `computeColumnSummaryStatistics`
4. `columnSimilarities`
5. `tallSkinnyQR` <sup>**[2]**</sup>
* `IndexedRowMatrix` <sup>**[3]**</sup>
1. `computeGramianMatrix`
* `CoordinateMatrix`
1. `transpose`
* `BlockMatrix`
1. `validate`
2. `cache`
3. `persist`
4. `transpose`
**[1]**: Note: `multiply`, `computeSVD`, and `computePrincipalComponents` are already part of PR #7963 for SPARK-6227.
**[2]**: Implementing `tallSkinnyQR` uncovered a bug with our PySpark `RowMatrix` constructor. As discussed on the dev list [here](http://apache-spark-developers-list.1001551.n3.nabble.com/K-Means-And-Class-Tags-td10038.html), there appears to be an issue with type erasure with RDDs coming from Java, and by extension from PySpark. Although we are attempting to construct a `RowMatrix` from an `RDD[Vector]` in [PythonMLlibAPI](https://github.com/apache/spark/blob/master/mllib/src/main/scala/org/apache/spark/mllib/api/python/PythonMLLibAPI.scala#L1115), the `Vector` type is erased, resulting in an `RDD[Object]`. Thus, when calling Scala's `tallSkinnyQR` from PySpark, we get a Java `ClassCastException` in which an `Object` cannot be cast to a Spark `Vector`. As noted in the aforementioned dev list thread, this issue was also encountered with `DecisionTrees`, and the fix involved an explicit `retag` of the RDD with a `Vector` type. Thus, this PR currently contains that fix applied to the `createRowMatrix` helper function in `PythonMLlibAPI`. `IndexedRowMatrix` and `CoordinateMatrix` do not appear to have this issue likely due to their related helper functions in `PythonMLlibAPI` creating the RDDs explicitly from DataFrames with pattern matching, thus preserving the types. However, this fix may be out of scope for this single PR, and it may be better suited in a separate JIRA/PR. Therefore, I have marked this PR as WIP and am open to discussion.
**[3]**: Note: `multiply` and `computeSVD` are already part of PR #7963 for SPARK-6227.
Author: Mike Dusenberry <mwdusenb@us.ibm.com>
Closes#9441 from dusenberrymw/SPARK-9656_Add_Missing_Methods_to_PySpark_Distributed_Linear_Algebra.
## What changes were proposed in this pull request?
Before, spark.ml GaussianMixtureModel used the spark.mllib MultivariateGaussian in its public API. This was added after 1.6, so we can modify this API without breaking APIs.
This PR copies MultivariateGaussian to mllib-local in spark.ml, with a few changes:
* Renamed fields to match numpy, scipy: mu => mean, sigma => cov
This PR then uses the spark.ml MultivariateGaussian in the spark.ml GaussianMixtureModel, which involves:
* Modifying the constructor
* Adding a computeProbabilities method
Also:
* Added EPSILON to mllib-local for use in MultivariateGaussian
## How was this patch tested?
Existing unit tests
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12593 from jkbradley/sparkml-gmm-fix.
## What changes were proposed in this pull request?
There have been continuing requests (e.g., SPARK-7131) for allowing users to extend and modify MLlib models and algorithms.
This PR makes tree and ensemble classes, Node types, and Split types in spark.ml no longer final. This matches most other spark.ml algorithms.
Constructors for models are still private since we may need to refactor how stats are maintained in tree nodes.
## How was this patch tested?
Existing unit tests
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12711 from jkbradley/final-trees.
## What changes were proposed in this pull request?
This PR changes `GLMRegressionModel.save` function like the following code that is similar to other algorithms' parquet write.
```
- val dataRDD: DataFrame = sc.parallelize(Seq(data), 1).toDF()
- // TODO: repartition with 1 partition after SPARK-5532 gets fixed
- dataRDD.write.parquet(Loader.dataPath(path))
+ sqlContext.createDataFrame(Seq(data)).repartition(1).write.parquet(Loader.dataPath(path))
```
## How was this patch tested?
Manual.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#12676 from dongjoon-hyun/SPARK-14907.
## What changes were proposed in this pull request?
We deprecated ```runs``` of mllib.KMeans in Spark 1.6 (SPARK-11358). In 2.0, we will make it no effect (with warning messages). We did not remove ```setRuns/getRuns``` for better binary compatibility.
This PR change `runs` which are appeared at the public API. Usage inside of ```KMeans.runAlgorithm()``` will be resolved at #10806.
## How was this patch tested?
Existing unit tests.
cc jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12608 from yanboliang/spark-11559.
## What changes were proposed in this pull request?
That patch mistakenly widened the visibility from `private[x]` to `protected[x]`. This patch reverts those changes.
Author: Andrew Or <andrew@databricks.com>
Closes#12686 from andrewor14/visibility.
## What changes were proposed in this pull request?
We currently have no way for users to propagate options to the underlying library that rely in Hadoop configurations to work. For example, there are various options in parquet-mr that users might want to set, but the data source API does not expose a per-job way to set it. This patch propagates the user-specified options also into Hadoop Configuration.
## How was this patch tested?
Used a mock data source implementation to test both the read path and the write path.
Author: Reynold Xin <rxin@databricks.com>
Closes#12688 from rxin/SPARK-14912.
## What changes were proposed in this pull request?
```AFTSurvivalRegressionModel``` supports ```save/load``` in SparkR.
## How was this patch tested?
Unit tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12685 from yanboliang/spark-14313.
## What changes were proposed in this pull request?
Made BinaryClassificationEvaluator, MulticlassClassificationEvaluator and RegressionEvaluator accept all numeric types for label
## How was this patch tested?
Unit tests
Author: BenFradet <benjamin.fradet@gmail.com>
Closes#12500 from BenFradet/SPARK-13962.
## What changes were proposed in this pull request?
In Spark 2.0, `SparkSession` is the new thing. Internally we should stop using `SQLContext` everywhere since that's supposed to be not the main user-facing API anymore.
In this patch I took care to not break any public APIs. The one place that's suspect is `o.a.s.ml.source.libsvm.DefaultSource`, but according to mengxr it's not supposed to be public so it's OK to change the underlying `FileFormat` trait.
**Reviewers**: This is a big patch that may be difficult to review but the changes are actually really straightforward. If you prefer I can break it up into a few smaller patches, but it will delay the progress of this issue a little.
## How was this patch tested?
No change in functionality intended.
Author: Andrew Or <andrew@databricks.com>
Closes#12625 from andrewor14/spark-session-refactor.
## What changes were proposed in this pull request?
SparkR ```NaiveBayesModel``` supports ```save/load``` by the following API:
```
df <- createDataFrame(sqlContext, infert)
model <- naiveBayes(education ~ ., df, laplace = 0)
ml.save(model, path)
model2 <- ml.load(path)
```
## How was this patch tested?
Add unit tests.
cc mengxr
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12573 from yanboliang/spark-14312.
## What changes were proposed in this pull request?
As the discussion at [SPARK-10574](https://issues.apache.org/jira/browse/SPARK-10574), ```HashingTF``` should support MurmurHash3 and make it as the default hash algorithm. We should also expose set/get API for ```hashAlgorithm```, then users can choose the hash method.
Note: The problem that ```mllib.feature.HashingTF``` behaves differently between Scala/Java and Python will be resolved in the followup work.
## How was this patch tested?
unit tests.
cc jkbradley MLnick
Author: Yanbo Liang <ybliang8@gmail.com>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12498 from yanboliang/spark-10574.
## What changes were proposed in this pull request?
Add Python API in ML for GaussianMixture
## How was this patch tested?
(Please explain how this patch was tested. E.g. unit tests, integration tests, manual tests)
Add doctest and test cases are the same as mllib Python tests
./dev/lint-python
PEP8 checks passed.
rm -rf _build/*
pydoc checks passed.
./python/run-tests --python-executables=python2.7 --modules=pyspark-ml
Running PySpark tests. Output is in /Users/mwang/spark_ws_0904/python/unit-tests.log
Will test against the following Python executables: ['python2.7']
Will test the following Python modules: ['pyspark-ml']
Finished test(python2.7): pyspark.ml.evaluation (18s)
Finished test(python2.7): pyspark.ml.clustering (40s)
Finished test(python2.7): pyspark.ml.classification (49s)
Finished test(python2.7): pyspark.ml.recommendation (44s)
Finished test(python2.7): pyspark.ml.feature (64s)
Finished test(python2.7): pyspark.ml.regression (45s)
Finished test(python2.7): pyspark.ml.tuning (30s)
Finished test(python2.7): pyspark.ml.tests (56s)
Tests passed in 106 seconds
Author: wm624@hotmail.com <wm624@hotmail.com>
Closes#12402 from wangmiao1981/gmm.
## What changes were proposed in this pull request?
add the checking for StepSize and Tol in sharedParams
## How was this patch tested?
Unit tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#12530 from zhengruifeng/ml_args_checking.
## What changes were proposed in this pull request?
Spark uses `NewLineAtEofChecker` rule in Scala by ScalaStyle. And, most Java code also comply with the rule. This PR aims to enforce the same rule `NewlineAtEndOfFile` by CheckStyle explicitly. Also, this fixes lint-java errors since SPARK-14465. The followings are the items.
- Adds a new line at the end of the files (19 files)
- Fixes 25 lint-java errors (12 RedundantModifier, 6 **ArrayTypeStyle**, 2 LineLength, 2 UnusedImports, 2 RegexpSingleline, 1 ModifierOrder)
## How was this patch tested?
After the Jenkins test succeeds, `dev/lint-java` should pass. (Currently, Jenkins dose not run lint-java.)
```bash
$ dev/lint-java
Using `mvn` from path: /usr/local/bin/mvn
Checkstyle checks passed.
```
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#12632 from dongjoon-hyun/SPARK-14868.
## What changes were proposed in this pull request?
del unused imports in ML/MLLIB
## How was this patch tested?
unit tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#12497 from zhengruifeng/del_unused_imports.
## What changes were proposed in this pull request?
We use `RowEncoder` in libsvm data source to serialize the label and features read from libsvm files. However, the schema passed in this encoder is not correct. As the result, we can't correctly select `features` column from the DataFrame. We should use full data schema instead of `requiredSchema` to serialize the data read in. Then do projection to select required columns later.
## How was this patch tested?
`LibSVMRelationSuite`.
Author: Liang-Chi Hsieh <simonh@tw.ibm.com>
Closes#12611 from viirya/fix-libsvm.
## What changes were proposed in this pull request?
1, fix the indentation
2, add a missing param desc
## How was this patch tested?
unit tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#12499 from zhengruifeng/fix_doc.
## What changes were proposed in this pull request?
Implement some `hashCode` and `equals` together in order to enable the scalastyle.
This is a first batch, I will continue to implement them but I wanted to know your thoughts.
Author: Joan <joan@goyeau.com>
Closes#12157 from joan38/SPARK-6429-HashCode-Equals.
## What changes were proposed in this pull request?
GLM supports output link prediction.
## How was this patch tested?
unit test.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12287 from yanboliang/spark-14479.
## What changes were proposed in this pull request?
For maintaining wrappers around spark.mllib algorithms in spark.ml, it will be useful to have ```private[spark]``` methods for converting from one linear algebra representation to another.
This PR adds toNew, fromNew methods for all spark.mllib Vector and Matrix types.
## How was this patch tested?
Unit tests for all conversions
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12504 from jkbradley/linalg-conversions.
## What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-14569
Log instrumentation in KMeans:
- featuresCol
- predictionCol
- k
- initMode
- initSteps
- maxIter
- seed
- tol
- summary
## How was this patch tested?
Manually test on local machine, by running and checking output of org.apache.spark.examples.ml.KMeansExample
Author: Xin Ren <iamshrek@126.com>
Closes#12432 from keypointt/SPARK-14569.
## What changes were proposed in this pull request?
Currently, MLlib's StandardScaler scales columns using the corrected standard deviation (sqrt of unbiased variance). This matches what R's scale package does.
This PR documents this fact.
## How was this patch tested?
doc only
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12519 from jkbradley/scaler-variance-doc.
## What changes were proposed in this pull request?
- replaced `FileSystem.get(conf)` calls with `path.getFileSystem(conf)`
## How was this patch tested?
N/A
Author: Liwei Lin <lwlin7@gmail.com>
Closes#12450 from lw-lin/fix-fs-get.
## What changes were proposed in this pull request?
This PR moves `HadoopFsRelation` related data source API into `execution/datasources` package.
Note that to avoid conflicts, this PR is based on #12153. Effective changes for this PR only consist of the last three commits. Will rebase after merging #12153.
## How was this patch tested?
Existing tests.
Author: Yin Huai <yhuai@databricks.com>
Author: Cheng Lian <lian@databricks.com>
Closes#12361 from liancheng/spark-14407-hide-hadoop-fs-relation.
## What changes were proposed in this pull request?
Added windowSize getter/setter to ML/MLlib
## How was this patch tested?
Added test cases in tests.py under both ML and MLlib
Author: Jason Lee <cjlee@us.ibm.com>
Closes#12428 from jasoncl/SPARK-14564.
## What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-14306
Add PySpark OneVsRest save/load supports.
## How was this patch tested?
Test with Python unit test.
Author: Xusen Yin <yinxusen@gmail.com>
Closes#12439 from yinxusen/SPARK-14306-0415.
## What changes were proposed in this pull request?
This PR removes
- Inappropriate type notations
For example, from
```scala
words.foreachRDD { (rdd: RDD[String], time: Time) =>
...
```
to
```scala
words.foreachRDD { (rdd, time) =>
...
```
- Extra anonymous closure within functional transformations.
For example,
```scala
.map(item => {
...
})
```
which can be just simply as below:
```scala
.map { item =>
...
}
```
and corrects some obvious style nits.
## How was this patch tested?
This was tested after adding rules in `scalastyle-config.xml`, which ended up with not finding all perfectly.
The rules applied were below:
- For the first correction,
```xml
<check customId="NoExtraClosure" level="error" class="org.scalastyle.file.RegexChecker" enabled="true">
<parameters><parameter name="regex">(?m)\.[a-zA-Z_][a-zA-Z0-9]*\(\s*[^,]+s*=>\s*\{[^\}]+\}\s*\)</parameter></parameters>
</check>
```
```xml
<check customId="NoExtraClosure" level="error" class="org.scalastyle.file.RegexChecker" enabled="true">
<parameters><parameter name="regex">\.[a-zA-Z_][a-zA-Z0-9]*\s*[\{|\(]([^\n>,]+=>)?\s*\{([^()]|(?R))*\}^[,]</parameter></parameters>
</check>
```
- For the second correction
```xml
<check customId="TypeNotation" level="error" class="org.scalastyle.file.RegexChecker" enabled="true">
<parameters><parameter name="regex">\.[a-zA-Z_][a-zA-Z0-9]*\s*[\{|\(]\s*\([^):]*:R))*\}^[,]</parameter></parameters>
</check>
```
**Those rules were not added**
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#12413 from HyukjinKwon/SPARK-style.
## What changes were proposed in this pull request?
Expose R-like summary statistics in SparkR::glm for more family and link functions.
Note: Not all values in R [summary.glm](http://stat.ethz.ch/R-manual/R-patched/library/stats/html/summary.glm.html) are exposed, we only provide the most commonly used statistics in this PR. More statistics can be added in the followup work.
## How was this patch tested?
Unit tests.
SparkR Output:
```
Deviance Residuals:
(Note: These are approximate quantiles with relative error <= 0.01)
Min 1Q Median 3Q Max
-0.95096 -0.16585 -0.00232 0.17410 0.72918
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.6765 0.23536 7.1231 4.4561e-11
Sepal_Length 0.34988 0.046301 7.5566 4.1873e-12
Species_versicolor -0.98339 0.072075 -13.644 0
Species_virginica -1.0075 0.093306 -10.798 0
(Dispersion parameter for gaussian family taken to be 0.08351462)
Null deviance: 28.307 on 149 degrees of freedom
Residual deviance: 12.193 on 146 degrees of freedom
AIC: 59.22
Number of Fisher Scoring iterations: 1
```
R output:
```
Deviance Residuals:
Min 1Q Median 3Q Max
-0.95096 -0.16522 0.00171 0.18416 0.72918
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.67650 0.23536 7.123 4.46e-11 ***
Sepal.Length 0.34988 0.04630 7.557 4.19e-12 ***
Speciesversicolor -0.98339 0.07207 -13.644 < 2e-16 ***
Speciesvirginica -1.00751 0.09331 -10.798 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for gaussian family taken to be 0.08351462)
Null deviance: 28.307 on 149 degrees of freedom
Residual deviance: 12.193 on 146 degrees of freedom
AIC: 59.217
Number of Fisher Scoring iterations: 2
```
cc mengxr
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12393 from yanboliang/spark-13925.
## What changes were proposed in this pull request?
Removed duplicated generation of `ids` in OnlineLDAOptimizer.
## How was this patch tested?
tested with existing unit tests.
Author: Pravin Gadakh <prgadakh@in.ibm.com>
Closes#12176 from pravingadakh/SPARK-14370.
## What changes were proposed in this pull request?
This task will copy the Vector and Matrix classes from mllib to ml package in mllib-local jar. The UDTs and `since` annotation in ml vector and matrix will be removed from now. UDTs will be achieved by #SPARK-14487, and `since` will be replaced by /* since 1.2.0 */
The BLAS implementation will be copied, and some of the test utilities will be copies as well.
Summary of changes:
1. In mllib-local/src/main/scala/org/apache/spark/**ml**/linalg/BLAS.scala
- Copied from mllib/src/main/scala/org/apache/spark/**mllib**/linalg/BLAS.scala
- logDebug("gemm: alpha is equal to 0 and beta is equal to 1. Returning C.") is removed in ml version.
2. In mllib-local/src/main/scala/org/apache/spark/**ml**/linalg/Matrices.scala
- Copied from mllib/src/main/scala/org/apache/spark/**mllib**/linalg/Matrices.scala
- `Since` was removed, and we'll use standard `/* Since /*` Java doc. Will be in another PR.
- `UDT` related code was removed, and will use `SPARK-13944` https://github.com/apache/spark/pull/12259 to replace the annotation.
3. In mllib-local/src/main/scala/org/apache/spark/**ml**/linalg/Vectors.scala
- Copied from mllib/src/main/scala/org/apache/spark/**mllib**/linalg/Vectors.scala
- `Since` was removed.
- `UDT` related code was removed.
- In `def parseNumeric`, it was throwing `throw new SparkException(s"Cannot parse $other.")`, and now it's throwing `throw new IllegalArgumentException(s"Cannot parse $other.")`
4. In mllib/src/main/scala/org/apache/spark/**mllib**/linalg/Vectors.scala
- For consistency with ML version of vector, `def parseNumeric` is now throwing `throw new IllegalArgumentException(s"Cannot parse $other.")`
5. mllib/src/main/scala/org/apache/spark/**mllib**/util/NumericParser.scala is moved to mllib-local/src/main/scala/org/apache/spark/**ml**/util/NumericParser.scala
- All the `throw new SparkException` were replaced by `throw new IllegalArgumentException`
## How was this patch tested?
unit tests
Author: DB Tsai <dbt@netflix.com>
Closes#12317 from dbtsai/dbtsai-ml-vector.
Hi guys,
I've implemented an improved version of the `toIndexedRowMatrix` function on the `BlockMatrix`. I needed this for a project, but would like to share it with the rest of the community. In the case of dense matrices, it can increase performance up to 19 times:
https://github.com/Fokko/BlockMatrixToIndexedRowMatrix
If there are any questions or suggestions, please let me know. Keep up the good work! Cheers.
Author: Fokko Driesprong <f.driesprong@catawiki.nl>
Author: Fokko Driesprong <fokko@driesprongen.nl>
Closes#10839 from Fokko/master.
## What changes were proposed in this pull request?
This fix tries to change RandomForest's supported strategies from using regexes to using parseInt and
parseDouble, for the purpose of robustness and maintainability.
## How was this patch tested?
Existing tests passed.
Author: Yong Tang <yong.tang.github@outlook.com>
Closes#12360 from yongtang/SPARK-14565.
## What changes were proposed in this pull request?
In Spark 1.4, we negated some metrics from RegressionEvaluator since CrossValidator always maximized metrics. This was fixed in 1.5, but the docs were not updated. This PR updates the docs.
## How was this patch tested?
no tests
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12377 from jkbradley/regeval-doc.
## What changes were proposed in this pull request?
Move json4s, breeze dependency declaration into parent
## How was this patch tested?
Should be no functional change, but Jenkins tests will test that.
Author: Sean Owen <sowen@cloudera.com>
Closes#12390 from srowen/SPARK-14612.
## What changes were proposed in this pull request?
* Modify ```KMeansSummary.clusterSizes``` method to make it robust to empty clusters.
* Add unit test for spark.ml ```KMeansSummary```.
* Add Since tag.
## How was this patch tested?
unit tests.
cc jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12254 from yanboliang/spark-14375.
## What changes were proposed in this pull request?
GLM training summaries should provide solver.
## How was this patch tested?
Unit tests.
cc jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12253 from yanboliang/spark-14461.
```PrefixSpanModel``` supports ```save/load```. It's similar with #9267.
cc jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#10664 from yanboliang/spark-10386.
## What changes were proposed in this pull request?
* Added save/load for ```GBTClassifier/GBTClassificationModel/GBTRegressor/GBTRegressionModel```.
* Meanwhile, I modified ```EnsembleModelReadWrite.saveImpl/loadImpl``` to support save/load ```treeWeights```.
## How was this patch tested?
Adds standard unit tests for GBT save/load.
cc jkbradley GayathriMurali
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12230 from yanboliang/spark-13783.
## What changes were proposed in this pull request?
This adds extra logging information about a `LogisticRegression` estimator when being fit on a dataset. With this PR, you see the following extra lines when running the example in the documentation:
```
16/04/13 07:19:00 INFO Instrumentation: Instrumentation(LogisticRegression-logreg_55dd3c09f164-1230977381-1): training: numPartitions=1 storageLevel=StorageLevel(disk=true, memory=true, offheap=false, deserialized=true, replication=1)
16/04/13 07:19:00 INFO Instrumentation: Instrumentation(LogisticRegression-logreg_55dd3c09f164-1230977381-1): {"regParam":0.3,"elasticNetParam":0.8,"maxIter":10}
...
16/04/12 11:48:07 INFO Instrumentation: Instrumentation(LogisticRegression-logreg_a89eb23cb386-358781145):numClasses=2
16/04/12 11:48:07 INFO Instrumentation: Instrumentation(LogisticRegression-logreg_a89eb23cb386-358781145):numFeatures=692
...
16/04/13 07:19:01 INFO Instrumentation: Instrumentation(LogisticRegression-logreg_55dd3c09f164-1230977381-1): training finished
```
## How was this patch tested?
This PR was manually tested.
Author: Timothy Hunter <timhunter@databricks.com>
Closes#12331 from thunterdb/1604-instrumentation.
## What changes were proposed in this pull request?
It looks several recent commits for datasources (maybe while removing old `HadoopFsRelation` interface) missed removing some unused imports.
This PR removes some unused imports in datasources.
## How was this patch tested?
`sbt scalastyle` and some unit tests for them.
Author: hyukjinkwon <gurwls223@gmail.com>
Closes#12326 from HyukjinKwon/minor-imports.
## What changes were proposed in this pull request?
SparkR does not support type of vector which is the default type of feature column in ML. R predict also does not output intermediate feature column. So SparkR ```predict``` should not output feature column. In this PR, I only fix this issue for ```naiveBayes``` and ```survreg```. ```kmeans``` has the right code route already and ```glm``` will be fixed at SparkRWrapper refactor(#12294).
## How was this patch tested?
No new tests.
cc mengxr shivaram
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11958 from yanboliang/spark-14147.
## What changes were proposed in this pull request?
Use a random table name instead of `__THIS__` in SQLTransformer, and add a test for `transformSchema`. The problems of using `__THIS__` are:
* It doesn't work under HiveContext (in Spark 1.6)
* Race conditions
## How was this patch tested?
* Manual test with HiveContext.
* Added a unit test for `transformSchema` to improve coverage.
cc: yhuai
Author: Xiangrui Meng <meng@databricks.com>
Closes#12330 from mengxr/SPARK-14563.
## What changes were proposed in this pull request?
AFTSurvivalRegression should support feature standardization, it will improve the convergence rate.
Test the convergence rate on the [Ovarian](https://stat.ethz.ch/R-manual/R-devel/library/survival/html/ovarian.html) data which is standard data comes with Survival library in R,
* without standardization(before this PR) -> 74 iterations.
* with standardization(after this PR) -> 38 iterations.
But after this fix, with or without ```standardization``` will converge to the same solution. It means that ```standardization = false``` will run the same code route as ```standardization = true```. Because if the features are not standardized at all, it will result convergency issue when the features have very different scales. This behavior is the same as ML [```LinearRegression``` and ```LogisticRegression```](https://issues.apache.org/jira/browse/SPARK-8522). See more discussion about this topic at #11247.
cc mengxr
## How was this patch tested?
unit test.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11365 from yanboliang/spark-13322.
* SparkR glm supports families and link functions which match R's signature for family.
* SparkR glm API refactor. The comparative standard of the new API is R glm, so I only expose the arguments that R glm supports: ```formula, family, data, epsilon and maxit```.
* This PR is focus on glm() and predict(), summary statistics will be done in a separate PR after this get in.
* This PR depends on #12287 which make GLMs support link prediction at Scala side. After that merged, I will add more tests for predict() to this PR.
Unit tests.
cc mengxr jkbradley hhbyyh
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12294 from yanboliang/spark-12566.
## What changes were proposed in this pull request?
This PR tries to support more options for feature subset size in RandomForest implementation. Previously, RandomForest only support "auto", "all", "sort", "log2", "onethird". This PR tries to support any given value to allow model search.
In this PR, `featureSubsetStrategy` could be passed with:
a) a real number in the range of `(0.0-1.0]` that represents the fraction of the number of features in each subset,
b) an integer number (`>0`) that represents the number of features in each subset.
## How was this patch tested?
Two tests `JavaRandomForestClassifierSuite` and `JavaRandomForestRegressorSuite` have been updated to check the additional options for params in this PR.
An additional test has been added to `org.apache.spark.mllib.tree.RandomForestSuite` to cover the cases in this PR.
Author: Yong Tang <yong.tang.github@outlook.com>
Closes#11989 from yongtang/SPARK-3724.
## What changes were proposed in this pull request?
According to the [Spark Code Style Guide](https://cwiki.apache.org/confluence/display/SPARK/Spark+Code+Style+Guide) and [Scala Style Guide](http://docs.scala-lang.org/style/control-structures.html#curlybraces), we had better enforce the following rule.
```
case: Always omit braces in case clauses.
```
This PR makes a new ScalaStyle rule, 'OmitBracesInCase', and enforces it to the code.
## How was this patch tested?
Pass the Jenkins tests (including Scala style checking)
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#12280 from dongjoon-hyun/SPARK-14508.
## What changes were proposed in this pull request?
Now `HadoopFsRelation` with all kinds of file formats can be handled in `FileSourceStrategy`, we can remove the branches for `HadoopFsRelation` in `FileSourceStrategy` and the `buildInternalScan` API from `FileFormat`.
## How was this patch tested?
existing tests.
Author: Wenchen Fan <wenchen@databricks.com>
Closes#12300 from cloud-fan/remove.
## What changes were proposed in this pull request?
Fixes to eliminate warnings during package and doc builds.
## How was this patch tested?
Existing unit tests
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12263 from jkbradley/warning-cleanups.
## What changes were proposed in this pull request?
This is follow up for #12089, add unit test for EM LDA which test disable checkpointing when set ```checkpointInterval = -1```.
## How was this patch tested?
unit test.
cc jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12286 from yanboliang/spark-14298-followup.
## What changes were proposed in this pull request?
QuantileDiscretizer can return an unexpected number of buckets in certain cases. This PR proposes to fix this issue and also refactor QuantileDiscretizer to use approxQuantiles from DataFrame stats functions.
## How was this patch tested?
QuantileDiscretizerSuite unit tests (some existing tests will change or even be removed in this PR)
Author: Oliver Pierson <ocp@gatech.edu>
Closes#11553 from oliverpierson/SPARK-13600.
## What changes were proposed in this pull request?
In order to separate the linear algebra, and vector matrix classes into a standalone jar, we need to setup the build first. This PR will create a new jar called mllib-local with minimal dependencies.
The previous PR was failing the build because of `spark-core:test` dependency, and that was reverted. In this PR, `FunSuite` with `// scalastyle:ignore funsuite` in mllib-local test was used, similar to sketch.
Thanks.
## How was this patch tested?
Unit tests
mengxr tedyu holdenk
Author: DB Tsai <dbt@netflix.com>
Closes#12298 from dbtsai/dbtsai-mllib-local-build-fix.
## What changes were proposed in this pull request?
add the checking for LDA and StreamingKMeans
## How was this patch tested?
manual tests
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#12062 from zhengruifeng/initmodel.
## What changes were proposed in this pull request?
This PR updates MLlib APIs to accept `Dataset[_]` as input where `DataFrame` was the input type. This PR doesn't change the output type. In Java, `Dataset[_]` maps to `Dataset<?>`, which includes `Dataset<Row>`. Some implementations were changed in order to return `DataFrame`. Tests and examples were updated. Note that this is a breaking change for subclasses of Transformer/Estimator.
Lol, we don't have to rename the input argument, which has been `dataset` since Spark 1.2.
TODOs:
- [x] update MiMaExcludes (seems all covered by explicit filters from SPARK-13920)
- [x] Python
- [x] add a new test to accept Dataset[LabeledPoint]
- [x] remove unused imports of Dataset
## How was this patch tested?
Exiting unit tests with some modifications.
cc: rxin jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#12274 from mengxr/SPARK-14500.
## What changes were proposed in this pull request?
Replace sortBy() with top() to calculate the top N frequent words as dictionary.
## How was this patch tested?
existing unit tests. The terms with same TF would be sorted in descending order. The test would fail if hardcode the terms with same TF the dictionary like "c", "d"...
Author: fwang1 <desperado.wf@gmail.com>
Closes#12265 from lionelfeng/master.
## What changes were proposed in this pull request?
In order to separate the linear algebra, and vector matrix classes into a standalone jar, we need to setup the build first. This PR will create a new jar called mllib-local with minimal dependencies. The test scope will still depend on spark-core and spark-core-test in order to use the common utilities, but the runtime will avoid any platform dependency. Couple platform independent classes will be moved to this package to demonstrate how this work.
## How was this patch tested?
Unit tests
Author: DB Tsai <dbt@netflix.com>
Closes#12241 from dbtsai/dbtsai-mllib-local-build.
## What changes were proposed in this pull request?
CountVectorizerModel has a binary toggle param. This PR is to add binary toggle param for estimator CountVectorizer. As discussed in the JIRA, instead of adding a param into CountVerctorizer, I moved the binary param to CountVectorizerParams. Therefore, the estimator inherits the binary param.
## How was this patch tested?
Add a new test case, which fits the model with binary flag set to true and then check the trained model's all non-zero counts is set to 1.0.
All tests in CounterVectorizerSuite.scala are passed.
Author: wm624@hotmail.com <wm624@hotmail.com>
Closes#12200 from wangmiao1981/binary_param.
## What changes were proposed in this pull request?
Cleanups to documentation. No changes to code.
* GBT docs: Move Scala doc for private object GradientBoostedTrees to public docs for GBTClassifier,Regressor
* GLM regParam: needs doc saying it is for L2 only
* TrainValidationSplitModel: add .. versionadded:: 2.0.0
* Rename “_transformer_params_from_java” to “_transfer_params_from_java”
* LogReg Summary classes: “probability” col should not say “calibrated”
* LR summaries: coefficientStandardErrors —> document that intercept stderr comes last. Same for t,p-values
* approxCountDistinct: Document meaning of “rsd" argument.
* LDA: note which params are for online LDA only
## How was this patch tested?
Doc build
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12266 from jkbradley/ml-doc-cleanups.
## What changes were proposed in this pull request?
In the doc of [```checkpointInterval```](https://github.com/apache/spark/blob/master/mllib/src/main/scala/org/apache/spark/ml/param/shared/sharedParams.scala#L241), we told users that they can disable checkpoint by setting ```checkpointInterval = -1```. But we did not handle this situation for LDA actually, we should fix this bug.
## How was this patch tested?
Existing tests.
cc jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12089 from yanboliang/spark-14298.
## What changes were proposed in this pull request?
The EMLDAOptimizer should generally not delete its last checkpoint since that can cause failures when DistributedLDAModel methods are called (if any partitions need to be recovered from the checkpoint).
This PR adds a "deleteLastCheckpoint" option which defaults to false. This is a change in behavior from Spark 1.6, in that the last checkpoint will not be removed by default.
This involves adding the deleteLastCheckpoint option to both spark.ml and spark.mllib, and modifying PeriodicCheckpointer to support the option.
This also:
* Makes MLlibTestSparkContext extend TempDirectory and set the checkpointDir to tempDir
* Updates LibSVMRelationSuite because of a name conflict with "tempDir" (and fixes a bug where it failed to delete a temp directory)
* Adds a MIMA exclude for DistributedLDAModel constructor, which is already ```private[clustering]```
## How was this patch tested?
Added 2 new unit tests to spark.ml LDASuite, which calls into spark.mllib.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12166 from jkbradley/emlda-save-checkpoint.
The current package name uses a dash, which is a little weird but seemed
to work. That is, until a new test tried to mock a class that references
one of those shaded types, and then things started failing.
Most changes are just noise to fix the logging configs.
For reference, SPARK-8815 also raised this issue, although at the time it
did not cause any issues in Spark, so it was not addressed.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#11941 from vanzin/SPARK-14134.
## What changes were proposed in this pull request?
This patch removes the implementation of gradient boosted trees in mllib/tree/GradientBoostedTrees.scala and changes mllib GBTs to call the implementation in spark.ML.
Primary changes:
* Removed `boost` method in mllib GradientBoostedTrees.scala
* Created new test suite GradientBoostedTreesSuite in ML, which contains unit tests that were specific to GBT internals from mllib
Other changes:
* Added an `updatePrediction` method in GradientBoostedTrees package. This method is added to provide consistency for methods that build predictions from boosted models. There are several methods that hard code the method of predicting as: sum_{i=1}^{numTrees} (treePrediction*treeWeight). Calling this function ensures that test methods that check accuracy use the same prediction method that the algorithm uses during training
* Added methods that were previously only used in testing, but were public methods, to GradientBoostedTrees. This includes `computeError` (previously part of `Loss` trait) and `evaluateEachIteration`. These are used in the new spark.ML unit tests. They are left in mllib as well so as to not break the API.
## How was this patch tested?
Existing unit tests which compare ML and MLlib ensure that mllib GBTs have not changed. Only a single unit test was moved to ML, which verifies that `runWithValidation` performs as expected.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#12050 from sethah/SPARK-12382.
## What changes were proposed in this pull request?
According to the [Spark Code Style Guide](https://cwiki.apache.org/confluence/display/SPARK/Spark+Code+Style+Guide#SparkCodeStyleGuide-Indentation), this PR adds a new scalastyle rule to prevent the followings.
```
/** In Spark, we don't use the ScalaDoc style so this
* is not correct.
*/
```
## How was this patch tested?
Pass the Jenkins tests (including `lint-scala`).
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#12221 from dongjoon-hyun/SPARK-14444.
## What changes were proposed in this pull request?
Adding Python API for training summaries of LogisticRegression and LinearRegression in PySpark ML.
## How was this patch tested?
Added unit tests to exercise the api calls for the summary classes. Also, manually verified values are expected and match those from Scala directly.
Author: Bryan Cutler <cutlerb@gmail.com>
Closes#11621 from BryanCutler/pyspark-ml-summary-SPARK-13430.
JIRA: https://issues.apache.org/jira/browse/SPARK-13538
## What changes were proposed in this pull request?
Add GaussianMixture and GaussianMixtureModel to ML package
## How was this patch tested?
unit tests and manual tests were done.
Local Scalastyle checks passed.
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Author: Ruifeng Zheng <ruifengz@foxmail.com>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#11419 from zhengruifeng/mlgmm.
## What changes were proposed in this pull request?
jira: https://issues.apache.org/jira/browse/SPARK-14322
OnlineLDAOptimizer uses RDD.reduce in two places where it could use treeAggregate. This can cause scalability issues. This should be an easy fix.
This is also a bug since it modifies the first argument to reduce, so we should use aggregate or treeAggregate.
See this line: f12f11e578/mllib/src/main/scala/org/apache/spark/mllib/clustering/LDAOptimizer.scala (L452)
and a few lines below it.
## How was this patch tested?
unit tests
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#12106 from hhbyyh/ldaTreeReduce.
## What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-13786
Add save/load for Python CrossValidator/Model and TrainValidationSplit/Model.
## How was this patch tested?
Test with Python doctest.
Author: Xusen Yin <yinxusen@gmail.com>
Closes#12020 from yinxusen/SPARK-13786.
## What changes were proposed in this pull request?
KMeansSummary class : deprecated size and added clusterSizes
Author: Shally Sangal <shallysangal@gmail.com>
Closes#12084 from shallys/master.
## What changes were proposed in this pull request?
In spark.ml, GBT and RandomForest expose the trait DecisionTreeModel in the trees method, but they should not since it is a private trait (and not ready to be made public). It will also be more useful to users if we return the concrete types.
This PR: return concrete types
The MIMA checks appear to be OK with this change.
## How was this patch tested?
Existing unit tests
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#12158 from jkbradley/hide-dtm.
## What changes were proposed in this pull request?
**Main change**: Added save/load for RandomForestClassifier, RandomForestRegressor (implementation details below)
Modified numTrees method (*deprecation*)
* Goal: Use default implementations of unit tests which assume Estimators and Models share the same set of Params.
* What this PR does: Moves method numTrees outside of trait TreeEnsembleModel. Adds it to GBT and RF Models. Deprecates it in RF Models in favor of new method getNumTrees. In Spark 2.1, we can have RF Models include Param numTrees.
Minor items
* Fixes bugs in GBTClassificationModel, GBTRegressionModel fromOld methods where they assign the wrong old UID.
**Implementation details**
* Split DecisionTreeModelReadWrite.loadTreeNodes into 2 methods in order to reuse some code for ensembles.
* Added EnsembleModelReadWrite object with save/load implementations usable for RFs and GBTs
* These store all trees' nodes in a single DataFrame, and all trees' metadata in a second DataFrame.
* Split trait RandomForestParams into parts in order to add more Estimator Params to RF models
* Split DefaultParamsWriter.saveMetadata into two methods to allow ensembles to store sub-models' metadata in a single DataFrame. Same for DefaultParamsReader.loadMetadata
## How was this patch tested?
Adds standard unit tests for RF save/load
Author: Joseph K. Bradley <joseph@databricks.com>
Author: GayathriMurali <gayathri.m.softie@gmail.com>
Closes#12118 from jkbradley/GayathriMurali-SPARK-13784.
## What changes were proposed in this pull request?
This PR contains the following 5 types of maintenance fix over 59 files (+94 lines, -93 lines).
- Fix typos(exception/log strings, testcase name, comments) in 44 lines.
- Fix lint-java errors (MaxLineLength) in 6 lines. (New codes after SPARK-14011)
- Use diamond operators in 40 lines. (New codes after SPARK-13702)
- Fix redundant semicolon in 5 lines.
- Rename class `InferSchemaSuite` to `CSVInferSchemaSuite` in CSVInferSchemaSuite.scala.
## How was this patch tested?
Manual and pass the Jenkins tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#12139 from dongjoon-hyun/SPARK-14355.
## What changes were proposed in this pull request?
This PR aims to fix all Scala-Style multiline comments into Java-Style multiline comments in Scala codes.
(All comment-only changes over 77 files: +786 lines, −747 lines)
## How was this patch tested?
Manual.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#12130 from dongjoon-hyun/use_multiine_javadoc_comments.
## What changes were proposed in this pull request?
Typo fixes. No functional changes.
## How was this patch tested?
Built the sources and ran with samples.
Author: Jacek Laskowski <jacek@japila.pl>
Closes#11802 from jaceklaskowski/typo-fixes.
## What changes were proposed in this pull request?
Decision tree helper classes will be migrated to ML. This patch moves those internal classes that are not part of the public API and removes ones that are no longer used, after [SPARK-12183](https://github.com/apache/spark/pull/11855). No functional changes are made.
Details:
* Bin.scala is removed as the ML implementation does not require bins
* mllib NodeIdCache is removed. It was only used by the mllib implementation previously, which no longer exists
* mllib TreePoint is removed. It was only used by the mllib implementation previously, which no longer exists
* BaggedPoint, DTStatsAggregator, DecisionTreeMetadata, BaggedPointSuite and TimeTracker are all moved to ML.
## How was this patch tested?
No functional changes are made. Existing unit tests ensure behavior is unchanged.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#12097 from sethah/cleanup_mllib_tree.
Currently, the Predictor abstraction expects the input labelCol type to be DoubleType, but we should support other numeric types. This will involve updating the PredictorParams.validateAndTransformSchema method.
Author: BenFradet <benjamin.fradet@gmail.com>
Closes#10355 from BenFradet/SPARK-7425.
## What changes were proposed in this pull request?
Fixes a compilation failure introduced in PR #12088 under Scala 2.10.
## How was this patch tested?
Compilation.
Author: Cheng Lian <lian@databricks.com>
Closes#12107 from liancheng/spark-14295-hotfix.
## What changes were proposed in this pull request?
Define and use ```KMeansWrapper``` for ```SparkR::kmeans```. It's only the code refactor for the original ```KMeans``` wrapper.
## How was this patch tested?
Existing tests.
cc mengxr
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12039 from yanboliang/spark-14059.
1.Implement LossFunction trait and implement squared error and cross entropy
loss with it
2.Implement unit test for gradient and loss
3.Implement InPlace trait and in-place layer evaluation
4.Refactor interface for ActivationFunction
5.Update of Layer and LayerModel interfaces
6.Fix random weights assignment
7.Implement memory allocation by MLP model instead of individual layers
These features decreased the memory usage and increased flexibility of
internal API.
Author: Alexander Ulanov <nashb@yandex.ru>
Author: avulanov <avulanov@gmail.com>
Closes#9229 from avulanov/mlp-refactoring.
## What changes were proposed in this pull request?
This PR implements `FileFormat.buildReader()` for the LibSVM data source. Besides that, a new interface method `prepareRead()` is added to `FileFormat`:
```scala
def prepareRead(
sqlContext: SQLContext,
options: Map[String, String],
files: Seq[FileStatus]): Map[String, String] = options
```
After migrating from `buildInternalScan()` to `buildReader()`, we lost the opportunity to collect necessary global information, since `buildReader()` works in a per-partition manner. For example, LibSVM needs to infer the total number of features if the `numFeatures` data source option is not set. Any necessary collected global information should be returned using the data source options map. By default, this method just returns the original options untouched.
An alternative approach is to absorb `inferSchema()` into `prepareRead()`, since schema inference is also some kind of global information gathering. However, this approach wasn't chosen because schema inference is optional, while `prepareRead()` must be called whenever a `HadoopFsRelation` based data source relation is instantiated.
One unaddressed problem is that, when `numFeatures` is absent, now the input data will be scanned twice. The `buildInternalScan()` code path doesn't need to do this because it caches the raw parsed RDD in memory before computing the total number of features. However, with `FileScanRDD`, the raw parsed RDD is created in a different way (e.g. partitioning) from the final RDD.
## How was this patch tested?
Tested using existing test suites.
Author: Cheng Lian <lian@databricks.com>
Closes#12088 from liancheng/spark-14295-libsvm-build-reader.
# What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-11892
Add save/load for spark ml.OneVsRest and its model. Also add OneVsRest and OneVsRestModel in MetaAlgorithmReadWrite.
# How was this patch tested?
Test with Scala unit test.
Author: Xusen Yin <yinxusen@gmail.com>
Closes#9934 from yinxusen/SPARK-11892.
## What changes were proposed in this pull request?
jira: https://issues.apache.org/jira/browse/SPARK-13782
Model export/import for BisectingKMeans in spark.ml and mllib
## How was this patch tested?
unit tests
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#11933 from hhbyyh/bisectingsave.
## What changes were proposed in this pull request?
This issue improves an input layer validation and adds related testcases to MultilayerPerceptronClassifier.
```scala
- // TODO: how to check ALSO that all elements are greater than 0?
- ParamValidators.arrayLengthGt(1)
+ (t: Array[Int]) => t.forall(ParamValidators.gt(0)) && t.length > 1
```
## How was this patch tested?
Pass the Jenkins tests including the new testcases.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11964 from dongjoon-hyun/SPARK-14164.
jira: https://issues.apache.org/jira/browse/SPARK-11507
"In certain situations when adding two block matrices, I get an error regarding colPtr and the operation fails. External issue URL includes full error and code for reproducing the problem."
root cause: colPtr.last does NOT always equal to values.length in breeze SCSMatrix, which fails the require in SparseMatrix.
easy step to repro:
```
val m1: BM[Double] = new CSCMatrix[Double] (Array (1.0, 1, 1), 3, 3, Array (0, 1, 2, 3), Array (0, 1, 2) )
val m2: BM[Double] = new CSCMatrix[Double] (Array (1.0, 2, 2, 4), 3, 3, Array (0, 0, 2, 4), Array (1, 2, 1, 2) )
val sum = m1 + m2
Matrices.fromBreeze(sum)
```
Solution: By checking the code in [CSCMatrix](28000a7b90/math/src/main/scala/breeze/linalg/CSCMatrix.scala), CSCMatrix in breeze can have extra zeros in the end of data array. Invoking compact will make sure it aligns with the require of SparseMatrix. This should add limited overhead as the actual compact operation is only performed when necessary.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#9520 from hhbyyh/matricesFromBreeze.
## What changes were proposed in this pull request?
Fix the wrong param name of LDA ```topicDistributionCol```.
## How was this patch tested?
No tests.
cc jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#12065 from yanboliang/lda-topicDistributionCol.
https://issues.apache.org/jira/browse/SPARK-14181
TrainValidationSplit should have HasSeed for the random split of RDD. I also changed the random split from the RDD function to the DataFrame function.
Author: Xusen Yin <yinxusen@gmail.com>
Closes#11985 from yinxusen/SPARK-14181.
## What changes were proposed in this pull request?
jira: https://issues.apache.org/jira/browse/SPARK-14154
I just read the code for KolmogorovSmirnovTest and find it could be much simplified following the original definition.
Send a PR for discussion
## How was this patch tested?
unit test
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#11954 from hhbyyh/ksoptimize.
## What changes were proposed in this pull request?
Adding binary toggle parameter to ml.feature.HashingTF, as well as mllib.feature.HashingTF since the former wraps this functionality. This parameter, if true, will set non-zero valued term counts to 1 to transform term count features to binary values that are well suited for discrete probability models.
## How was this patch tested?
Added unit tests for ML and MLlib
Author: Bryan Cutler <cutlerb@gmail.com>
Closes#11832 from BryanCutler/binary-param-HashingTF-SPARK-13963.
## What changes were proposed in this pull request?
Now that GBTs have been moved to ML, they can use the implementation of feature importance for random forests. This patch simply adds a `featureImportances` attribute to `GBTClassifier` and `GBTRegressor` and adds tests for each.
GBT feature importances here simply average the feature importances for each tree in its ensemble. This follows the implementation from scikit-learn. This method is also suggested by J Friedman in [this paper](https://statweb.stanford.edu/~jhf/ftp/trebst.pdf).
## How was this patch tested?
Unit tests were added to `GBTClassifierSuite` and `GBTRegressorSuite` to validate feature importances.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#11961 from sethah/SPARK-11730.
https://issues.apache.org/jira/browse/SPARK-11893
jkbradley In order to share read/write with `TrainValidationSplit`, I move the `SharedReadWrite` out of `CrossValidator` into a new trait `SharedReadWrite` in the tunning package.
To reduce the repeated tests, I move the complex tests from `CrossValidatorSuite` to `SharedReadWriteSuite`, and create a fake validator called `MyValidator` to test the shared code.
With `SharedReadWrite`, potential newly added `Validator` can share the read/write common part, and only need to implement their extra params save/load.
Author: Xusen Yin <yinxusen@gmail.com>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9971 from yinxusen/SPARK-11893.
## What changes were proposed in this pull request?
Fix incorrect use of binarySearch in SparseMatrix
## How was this patch tested?
Unit test added.
Author: Chenliang Xu <chexu@groupon.com>
Closes#11992 from luckyrandom/SPARK-14187.
## What changes were proposed in this pull request?
Better error message with k-means init can't be enough samples from input (because it is perhaps empty)
## How was this patch tested?
Jenkins tests.
Author: Sean Owen <sowen@cloudera.com>
Closes#11979 from srowen/SPARK-12494.
## What changes were proposed in this pull request?
Made evaluate method public. Fixed LogisticRegressionModel evaluate to handle case when probabilityCol is not specified.
## How was this patch tested?
There were already unit tests for these methods.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#11928 from jkbradley/public-evaluate.
## What changes were proposed in this pull request?
This PR fixes the following line and the related code. Historically, this code was added in [SPARK-5597](https://issues.apache.org/jira/browse/SPARK-5597). After [SPARK-5597](https://issues.apache.org/jira/browse/SPARK-5597) was committed, [SPARK-3365](https://issues.apache.org/jira/browse/SPARK-3365) is fixed now. Now, we had better remove the comment without changing persistent code.
```scala
- categories: Seq[Double]) { // TODO: Change to List once SPARK-3365 is fixed
+ categories: Seq[Double]) {
```
## How was this patch tested?
Pass the Jenkins tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11966 from dongjoon-hyun/change_categories_type.
## What changes were proposed in this pull request?
Removed methods that has been deprecated since 1.1, 1.2, 1.3, 1.4, and 1.5.
## How was this patch tested?
- manully checked that no codes in Spark call these methods any more
- existing test suits
Author: Liwei Lin <lwlin7@gmail.com>
Author: proflin <proflin.me@gmail.com>
Closes#11910 from lw-lin/remove-deprecates.
## What changes were proposed in this pull request?
StringIndexerModel.transform sets the output column metadata to use name inputCol. It should not. Fixing this causes a problem with the metadata produced by RFormula.
Fix in RFormula: I added the StringIndexer columns to prefixesToRewrite, and I modified VectorAttributeRewriter to find and replace all "prefixes" since attributes collect multiple prefixes from StringIndexer + Interaction.
Note that "prefixes" is no longer accurate since internal strings may be replaced.
## How was this patch tested?
Unit test which failed before this fix.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#11965 from jkbradley/StringIndexer-fix.
## What changes were proposed in this pull request?
This PR continues the work in #11447, we implemented the wrapper of ```AFTSurvivalRegression``` named ```survreg``` in SparkR.
## How was this patch tested?
Test against output from R package survival's survreg.
cc mengxr felixcheung
Close#11447
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11932 from yanboliang/spark-13010-new.
## What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-11871
Add save/load for MLPC
## How was this patch tested?
Test with Scala unit test
Author: Xusen Yin <yinxusen@gmail.com>
Closes#9854 from yinxusen/SPARK-11871.
## What changes were proposed in this pull request?
Just a typo
## How was this patch tested?
N/A
Author: Juarez Bochi <jbochi@gmail.com>
Closes#11896 from jbochi/patch-1.
Primary change:
* Removed spark.mllib.tree.DecisionTree implementation of tree and forest learning.
* spark.mllib now calls the spark.ml implementation.
* Moved unit tests (of tree learning internals) from spark.mllib to spark.ml as needed.
ml.tree.DecisionTreeModel
* Added toOld and made ```private[spark]```, implemented for Classifier and Regressor in subclasses. These methods now use OldInformationGainStats.invalidInformationGainStats for LeafNodes in order to mimic the spark.mllib implementation.
ml.tree.Node
* Added ```private[tree] def deepCopy```, used by unit tests
Copied developer comments from spark.mllib implementation to spark.ml one.
Moving unit tests
* Tree learning internals were tested by spark.mllib.tree.DecisionTreeSuite, or spark.mllib.tree.RandomForestSuite.
* Those tests were all moved to spark.ml.tree.impl.RandomForestSuite. The order in the file + the test names are the same, so you should be able to compare them by opening them in 2 windows side-by-side.
* I made minimal changes to each test to allow it to run. Each test makes the same checks as before, except for a few removed assertions which were checking irrelevant values.
* No new unit tests were added.
* mllib.tree.DecisionTreeSuite: I removed some checks of splits and bins which were not relevant to the unit tests they were in. Those same split calculations were already being tested in other unit tests, for each dataset type.
**Changes of behavior** (to be noted in SPARK-13448 once this PR is merged)
* spark.ml.tree.impl.RandomForest: Rather than throwing an error when maxMemoryInMB is set to too small a value (to split any node), we now allow 1 node to be split, even if its memory requirements exceed maxMemoryInMB. This involved removing the maxMemoryPerNode check in RandomForest.run, as well as modifying selectNodesToSplit(). Once this PR is merged, I will note the change of behavior on SPARK-13448.
* spark.mllib.tree.DecisionTree: When a tree only has one node (root = leaf node), the "stats" field will now be empty, rather than being set to InformationGainStats.invalidInformationGainStats. This does not remove information from the tree, and it will save a bit of storage.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#11855 from jkbradley/remove-mllib-tree-impl.
## What changes were proposed in this pull request?
`GBTClassifier` and `GBTRegressor` should use random seed for reproducible results. Because of the nature of current unit tests, which compare GBTs in ML and GBTs in MLlib for equality, I also added a random seed to MLlib GBT algorithm. I made alternate constructors in `mllib.tree.GradientBoostedTrees` to accept a random seed, but left them as private so as to not change the API unnecessarily.
## How was this patch tested?
Existing unit tests verify that functionality did not change. Other ML algorithms do not seem to have unit tests that directly test the functionality of random seeding, but reproducibility with seeding for GBTs is effectively verified in existing tests. I can add more tests if needed.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#11903 from sethah/SPARK-13952.
## What changes were proposed in this pull request?
Print more info about failed NaiveBayesSuite tests which have exhibited flakiness.
## How was this patch tested?
Ran locally with incorrect check to cause failure.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#11858 from jkbradley/naive-bayes-bug-log.
## What changes were proposed in this pull request?
This PR continues the work in #11486 from yinxusen with some code refactoring. In R package e1071, `naiveBayes` supports both categorical (Bernoulli) and continuous features (Gaussian), while in MLlib we support Bernoulli and multinomial. This PR implements the common subset: Bernoulli.
I moved the implementation out from SparkRWrappers to NaiveBayesWrapper to make it easier to read. Argument names, default values, and summary now match e1071's naiveBayes.
I removed the preprocess part that omit NA values because we don't know which columns to process.
## How was this patch tested?
Test against output from R package e1071's naiveBayes.
cc: yanboliang yinxusen
Closes#11486
Author: Xusen Yin <yinxusen@gmail.com>
Author: Xiangrui Meng <meng@databricks.com>
Closes#11890 from mengxr/SPARK-13449.
## What changes were proposed in this pull request?
Spark uses `DeveloperApi` annotation, but sometimes it seems to conflict with visibility. This PR tries to fix those conflict by removing annotations for non-publics. The following is the example.
**JobResult.scala**
```scala
DeveloperApi
sealed trait JobResult
DeveloperApi
case object JobSucceeded extends JobResult
-DeveloperApi
private[spark] case class JobFailed(exception: Exception) extends JobResult
```
## How was this patch tested?
Pass the existing Jenkins test.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11797 from dongjoon-hyun/SPARK-13986.
## What changes were proposed in this pull request?
[Spark Coding Style Guide](https://cwiki.apache.org/confluence/display/SPARK/Spark+Code+Style+Guide) has 100-character limit on lines, but it's disabled for Java since 11/09/15. This PR enables **LineLength** checkstyle again. To help that, this also introduces **RedundantImport** and **RedundantModifier**, too. The following is the diff on `checkstyle.xml`.
```xml
- <!-- TODO: 11/09/15 disabled - the lengths are currently > 100 in many places -->
- <!--
<module name="LineLength">
<property name="max" value="100"/>
<property name="ignorePattern" value="^package.*|^import.*|a href|href|http://|https://|ftp://"/>
</module>
- -->
<module name="NoLineWrap"/>
<module name="EmptyBlock">
<property name="option" value="TEXT"/>
-167,5 +164,7
</module>
<module name="CommentsIndentation"/>
<module name="UnusedImports"/>
+ <module name="RedundantImport"/>
+ <module name="RedundantModifier"/>
```
## How was this patch tested?
Currently, `lint-java` is disabled in Jenkins. It needs a manual test.
After passing the Jenkins tests, `dev/lint-java` should passes locally.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11831 from dongjoon-hyun/SPARK-14011.
This PR changes the `findSplits` method in spark.ml to perform split calculations on the workers. This PR is meant to copy [PR-8246](https://github.com/apache/spark/pull/8246) which added the same feature for MLlib.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#10231 from sethah/SPARK-12182.
## What changes were proposed in this pull request?
This is a continued work for https://github.com/apache/spark/pull/11536#issuecomment-198511013,
containing some comment update and style adjustment.
jkbradley
## How was this patch tested?
unit tests.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#11830 from hhbyyh/cvToggle.
## What changes were proposed in this pull request?
When trainingSummary is None, it should throw ```RuntimeException```.
cc mengxr
## How was this patch tested?
Existing tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11784 from yanboliang/fix-summary.
Decision trees in spark.ml (RandomForest.scala) communicate twice as much data as needed for unordered categorical features. Here's an example.
Say there are 3 categories A, B, C. We consider 3 splits:
* A vs. B, C
* A, B vs. C
* A, C vs. B
Currently, we collect statistics for each of the 6 subsets of categories (3 * 2 = 6). However, we could instead collect statistics for the 3 subsets on the left-hand side of the 3 possible splits: A and A,B and A,C. If we also have stats for the entire node, then we can compute the stats for the 3 subsets on the right-hand side of the splits. In pseudomath: stats(B,C) = stats(A,B,C) - stats(A).
This patch adds a parent stats array to the `DTStatsAggregator` so that the right child stats do not need to be stored. The right child stats are computed by subtracting left child stats from the parent stats for unordered categorical features.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#9474 from sethah/SPARK-10788.
## What changes were proposed in this pull request?
Cleanups from [https://github.com/apache/spark/pull/11620]: remove remaining uses of validateParams, and put functionality into transformSchema
## How was this patch tested?
Existing unit tests, modified to check using transformSchema instead of validateParams
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#11790 from jkbradley/SPARK-13761-cleanup.
## What changes were proposed in this pull request?
Logging was made private in Spark 2.0. If we move it, then users would be able to create a Logging trait themselves to avoid changing their own code.
## How was this patch tested?
existing tests.
Author: Wenchen Fan <wenchen@databricks.com>
Closes#11764 from cloud-fan/logger.
## What changes were proposed in this pull request?
Deprecate validateParams() method here: 035d3acdf3/mllib/src/main/scala/org/apache/spark/ml/param/params.scala (L553)
Move all functionality in overridden methods to transformSchema().
Check docs to make sure they indicate complex Param interaction checks should be done in transformSchema.
## How was this patch tested?
unit tests
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#11620 from hhbyyh/depreValid.
## What changes were proposed in this pull request?
Narrow down the parameter type of `UserDefinedType#serialize()`. Currently, the parameter type is `Any`, however it would logically make more sense to narrow it down to the type of the actual user defined type.
## How was this patch tested?
Existing tests were successfully run on local machine.
Author: Jakob Odersky <jakob@odersky.com>
Closes#11379 from jodersky/SPARK-11011-udt-types.
## What changes were proposed in this pull request?
Add row/column iterator to local matrices to simplify tasks like BlockMatrix => RowMatrix conversion. It handles dense and sparse matrices properly.
## How was this patch tested?
Unit tests on sparse and dense matrix.
cc: dbtsai
Author: Xiangrui Meng <meng@databricks.com>
Closes#11757 from mengxr/SPARK-13927.
### What changes were proposed in this pull request?
Made these MLReadable and MLWritable: DecisionTreeClassifier, DecisionTreeClassificationModel, DecisionTreeRegressor, DecisionTreeRegressionModel
* The shared implementation is in treeModels.scala
* I use case classes to create a DataFrame to save, and I use the Dataset API to parse loaded files.
Other changes:
* Made CategoricalSplit.numCategories public (to use in persistence)
* Fixed a bug in DefaultReadWriteTest.testEstimatorAndModelReadWrite, where it did not call the checkModelData function passed as an argument. This caused an error in LDASuite, which I fixed.
### How was this patch tested?
Persistence is tested via unit tests. For each algorithm, there are 2 non-trivial trees (depth 2). One is built with continuous features, and one with categorical; this ensures that both types of splits are tested.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#11581 from jkbradley/dt-io.
## What changes were proposed in this pull request?
Provide ignored test cases to export the test dataset into CSV format in ```LinearRegressionSuite```, ```LogisticRegressionSuite```, ```AFTSurvivalRegressionSuite``` and ```GeneralizedLinearRegressionSuite```, so users can validate the training accuracy compared with R's glm, glmnet and survival package.
cc mengxr
## How was this patch tested?
The test suite is ignored, but I have enabled all these cases offline and it works as expected.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11463 from yanboliang/spark-13613.
## What changes were proposed in this pull request?
https://issues.apache.org/jira/browse/SPARK-13894
Change the return type of the `SQLContext.range` API from `DataFrame` to `Dataset`.
## How was this patch tested?
No additional unit test required.
Author: Cheng Hao <hao.cheng@intel.com>
Closes#11730 from chenghao-intel/range.
## What changes were proposed in this pull request?
Follow up to https://github.com/apache/spark/pull/11657
- Also update `String.getBytes("UTF-8")` to use `StandardCharsets.UTF_8`
- And fix one last new Coverity warning that turned up (use of unguarded `wait()` replaced by simpler/more robust `java.util.concurrent` classes in tests)
- And while we're here cleaning up Coverity warnings, just fix about 15 more build warnings
## How was this patch tested?
Jenkins tests
Author: Sean Owen <sowen@cloudera.com>
Closes#11725 from srowen/SPARK-13823.2.
## What changes were proposed in this pull request?
Provide R-like summary statistics for GLMs via iteratively reweighted least squares.
## How was this patch tested?
unit tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11694 from yanboliang/spark-9837.
Currently, GBTs in spark.ml wrap the implementation in spark.mllib. This is preventing several improvements to GBTs in spark.ml, so we need to move the implementation to ml and use spark.ml decision trees in the implementation. At first, we should make minimal changes to the implementation.
Performance testing should be done to ensure there were no regressions.
Performance testing results are [here](https://docs.google.com/document/d/1dYd2mnfGdUKkQ3vZe2BpzsTnI5IrpSLQ-NNKDZhUkgw/edit?usp=sharing)
Author: sethah <seth.hendrickson16@gmail.com>
Closes#10607 from sethah/SPARK-12379.
This PR adds a new strategy, `FileSourceStrategy`, that can be used for planning scans of collections of files that might be partitioned or bucketed.
Compared with the existing planning logic in `DataSourceStrategy` this version has the following desirable properties:
- It removes the need to have `RDD`, `broadcastedHadoopConf` and other distributed concerns in the public API of `org.apache.spark.sql.sources.FileFormat`
- Partition column appending is delegated to the format to avoid an extra copy / devectorization when appending partition columns
- It minimizes the amount of data that is shipped to each executor (i.e. it does not send the whole list of files to every worker in the form of a hadoop conf)
- it natively supports bucketing files into partitions, and thus does not require coalescing / creating a `UnionRDD` with the correct partitioning.
- Small files are automatically coalesced into fewer tasks using an approximate bin-packing algorithm.
Currently only a testing source is planned / tested using this strategy. In follow-up PRs we will port the existing formats to this API.
A stub for `FileScanRDD` is also added, but most methods remain unimplemented.
Other minor cleanups:
- partition pruning is pushed into `FileCatalog` so both the new and old code paths can use this logic. This will also allow future implementations to use indexes or other tricks (i.e. a MySQL metastore)
- The partitions from the `FileCatalog` now propagate information about file sizes all the way up to the planner so we can intelligently spread files out.
- `Array` -> `Seq` in some internal APIs to avoid unnecessary `toArray` calls
- Rename `Partition` to `PartitionDirectory` to differentiate partitions used earlier in pruning from those where we have already enumerated the files and their sizes.
Author: Michael Armbrust <michael@databricks.com>
Closes#11646 from marmbrus/fileStrategy.
## What changes were proposed in this pull request?
`LinearRegressionWithSGD` and `StreamingLinearRegressionWithSGD` does not have `regParam` as their constructor arguments. They just depends on GradientDescent's default reqParam values.
To be consistent with other algorithms, we had better add them. The same default value is used.
## How was this patch tested?
Pass the existing unit test.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11527 from dongjoon-hyun/SPARK-13686.
## What changes were proposed in this pull request?
This PR fixes 135 typos over 107 files:
* 121 typos in comments
* 11 typos in testcase name
* 3 typos in log messages
## How was this patch tested?
Manual.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11689 from dongjoon-hyun/fix_more_typos.
## What changes were proposed in this pull request?
- Fixes calls to `new String(byte[])` or `String.getBytes()` that rely on platform default encoding, to use UTF-8
- Same for `InputStreamReader` and `OutputStreamWriter` constructors
- Standardizes on UTF-8 everywhere
- Standardizes specifying the encoding with `StandardCharsets.UTF-8`, not the Guava constant or "UTF-8" (which means handling `UnuspportedEncodingException`)
- (also addresses the other remaining Coverity scan issues, which are pretty trivial; these are separated into commit 1deecd8d9c )
## How was this patch tested?
Jenkins tests
Author: Sean Owen <sowen@cloudera.com>
Closes#11657 from srowen/SPARK-13823.
## What changes were proposed in this pull request?
This PR removes two methods, `collectRows()` and `takeRows()`, from `Dataset[T]`. These methods were added in PR #11443, and were later considered not useful.
## How was this patch tested?
Existing tests should do the work.
Author: Cheng Lian <lian@databricks.com>
Closes#11678 from liancheng/remove-collect-rows-and-take-rows.
## What changes were proposed in this pull request?
This PR unifies DataFrame and Dataset by migrating existing DataFrame operations to Dataset and make `DataFrame` a type alias of `Dataset[Row]`.
Most Scala code changes are source compatible, but Java API is broken as Java knows nothing about Scala type alias (mostly replacing `DataFrame` with `Dataset<Row>`).
There are several noticeable API changes related to those returning arrays:
1. `collect`/`take`
- Old APIs in class `DataFrame`:
```scala
def collect(): Array[Row]
def take(n: Int): Array[Row]
```
- New APIs in class `Dataset[T]`:
```scala
def collect(): Array[T]
def take(n: Int): Array[T]
def collectRows(): Array[Row]
def takeRows(n: Int): Array[Row]
```
Two specialized methods `collectRows` and `takeRows` are added because Java doesn't support returning generic arrays. Thus, for example, `DataFrame.collect(): Array[T]` actually returns `Object` instead of `Array<T>` from Java side.
Normally, Java users may fall back to `collectAsList` and `takeAsList`. The two new specialized versions are added to avoid performance regression in ML related code (but maybe I'm wrong and they are not necessary here).
1. `randomSplit`
- Old APIs in class `DataFrame`:
```scala
def randomSplit(weights: Array[Double], seed: Long): Array[DataFrame]
def randomSplit(weights: Array[Double]): Array[DataFrame]
```
- New APIs in class `Dataset[T]`:
```scala
def randomSplit(weights: Array[Double], seed: Long): Array[Dataset[T]]
def randomSplit(weights: Array[Double]): Array[Dataset[T]]
```
Similar problem as above, but hasn't been addressed for Java API yet. We can probably add `randomSplitAsList` to fix this one.
1. `groupBy`
Some original `DataFrame.groupBy` methods have conflicting signature with original `Dataset.groupBy` methods. To distinguish these two, typed `Dataset.groupBy` methods are renamed to `groupByKey`.
Other noticeable changes:
1. Dataset always do eager analysis now
We used to support disabling DataFrame eager analysis to help reporting partially analyzed malformed logical plan on analysis failure. However, Dataset encoders requires eager analysi during Dataset construction. To preserve the error reporting feature, `AnalysisException` now takes an extra `Option[LogicalPlan]` argument to hold the partially analyzed plan, so that we can check the plan tree when reporting test failures. This plan is passed by `QueryExecution.assertAnalyzed`.
## How was this patch tested?
Existing tests do the work.
## TODO
- [ ] Fix all tests
- [ ] Re-enable MiMA check
- [ ] Update ScalaDoc (`since`, `group`, and example code)
Author: Cheng Lian <lian@databricks.com>
Author: Yin Huai <yhuai@databricks.com>
Author: Wenchen Fan <wenchen@databricks.com>
Author: Cheng Lian <liancheng@users.noreply.github.com>
Closes#11443 from liancheng/ds-to-df.
## What changes were proposed in this pull request?
Since the opening curly brace, '{', has many usages as discussed in [SPARK-3854](https://issues.apache.org/jira/browse/SPARK-3854), this PR adds a ScalaStyle rule to prevent '){' pattern for the following majority pattern and fixes the code accordingly. If we enforce this in ScalaStyle from now, it will improve the Scala code quality and reduce review time.
```
// Correct:
if (true) {
println("Wow!")
}
// Incorrect:
if (true){
println("Wow!")
}
```
IntelliJ also shows new warnings based on this.
## How was this patch tested?
Pass the Jenkins ScalaStyle test.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11637 from dongjoon-hyun/SPARK-3854.
Adding support for other numeric types:
* Integer
* Short
* Long
* Float
* Decimal
Author: sethah <seth.hendrickson16@gmail.com>
Closes#9777 from sethah/SPARK-11108.
This patch adds an API entry point for single decision tree feature importances.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#9912 from sethah/SPARK-11861.
## What changes were proposed in this pull request?
```GeneralizedLinearRegression``` supports ```save/load```.
cc mengxr
## How was this patch tested?
unit test.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11465 from yanboliang/spark-13615.
## What changes were proposed in this pull request?
In order to make `docs/examples` (and other related code) more simple/readable/user-friendly, this PR replaces existing codes like the followings by using `diamond` operator.
```
- final ArrayList<Product2<Object, Object>> dataToWrite =
- new ArrayList<Product2<Object, Object>>();
+ final ArrayList<Product2<Object, Object>> dataToWrite = new ArrayList<>();
```
Java 7 or higher supports **diamond** operator which replaces the type arguments required to invoke the constructor of a generic class with an empty set of type parameters (<>). Currently, Spark Java code use mixed usage of this.
## How was this patch tested?
Manual.
Pass the existing tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11541 from dongjoon-hyun/SPARK-13702.
## What changes were proposed in this pull request?
Although we defined ```checkModelData``` in [```read/write``` test](https://github.com/apache/spark/blob/master/mllib/src/test/scala/org/apache/spark/ml/regression/LinearRegressionSuite.scala#L994) of ML estimators/models and pass it to ```testEstimatorAndModelReadWrite```, ```testEstimatorAndModelReadWrite``` omits to call ```checkModelData``` to check the equality of model data. So actually we did not run the check of model data equality for all test cases currently, we should fix it.
BTW, fix the bug of LDA read/write test which did not set ```docConcentration```. This bug should have failed test, but it does not complain because we did not run ```checkModelData``` actually.
cc jkbradley mengxr
## How was this patch tested?
No new unit test, should pass the exist ones.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11513 from yanboliang/ml-check-model-data.
## What changes were proposed in this pull request?
Remove last usage of jblas, in tests
## How was this patch tested?
Jenkins tests -- the same ones that are being modified.
Author: Sean Owen <sowen@cloudera.com>
Closes#11560 from srowen/SPARK-13715.
`HadoopFsRelation` is used for reading most files into Spark SQL. However today this class mixes the concerns of file management, schema reconciliation, scan building, bucketing, partitioning, and writing data. As a result, many data sources are forced to reimplement the same functionality and the various layers have accumulated a fair bit of inefficiency. This PR is a first cut at separating this into several components / interfaces that are each described below. Additionally, all implementations inside of Spark (parquet, csv, json, text, orc, svmlib) have been ported to the new API `FileFormat`. External libraries, such as spark-avro will also need to be ported to work with Spark 2.0.
### HadoopFsRelation
A simple `case class` that acts as a container for all of the metadata required to read from a datasource. All discovery, resolution and merging logic for schemas and partitions has been removed. This an internal representation that no longer needs to be exposed to developers.
```scala
case class HadoopFsRelation(
sqlContext: SQLContext,
location: FileCatalog,
partitionSchema: StructType,
dataSchema: StructType,
bucketSpec: Option[BucketSpec],
fileFormat: FileFormat,
options: Map[String, String]) extends BaseRelation
```
### FileFormat
The primary interface that will be implemented by each different format including external libraries. Implementors are responsible for reading a given format and converting it into `InternalRow` as well as writing out an `InternalRow`. A format can optionally return a schema that is inferred from a set of files.
```scala
trait FileFormat {
def inferSchema(
sqlContext: SQLContext,
options: Map[String, String],
files: Seq[FileStatus]): Option[StructType]
def prepareWrite(
sqlContext: SQLContext,
job: Job,
options: Map[String, String],
dataSchema: StructType): OutputWriterFactory
def buildInternalScan(
sqlContext: SQLContext,
dataSchema: StructType,
requiredColumns: Array[String],
filters: Array[Filter],
bucketSet: Option[BitSet],
inputFiles: Array[FileStatus],
broadcastedConf: Broadcast[SerializableConfiguration],
options: Map[String, String]): RDD[InternalRow]
}
```
The current interface is based on what was required to get all the tests passing again, but still mixes a couple of concerns (i.e. `bucketSet` is passed down to the scan instead of being resolved by the planner). Additionally, scans are still returning `RDD`s instead of iterators for single files. In a future PR, bucketing should be removed from this interface and the scan should be isolated to a single file.
### FileCatalog
This interface is used to list the files that make up a given relation, as well as handle directory based partitioning.
```scala
trait FileCatalog {
def paths: Seq[Path]
def partitionSpec(schema: Option[StructType]): PartitionSpec
def allFiles(): Seq[FileStatus]
def getStatus(path: Path): Array[FileStatus]
def refresh(): Unit
}
```
Currently there are two implementations:
- `HDFSFileCatalog` - based on code from the old `HadoopFsRelation`. Infers partitioning by recursive listing and caches this data for performance
- `HiveFileCatalog` - based on the above, but it uses the partition spec from the Hive Metastore.
### ResolvedDataSource
Produces a logical plan given the following description of a Data Source (which can come from DataFrameReader or a metastore):
- `paths: Seq[String] = Nil`
- `userSpecifiedSchema: Option[StructType] = None`
- `partitionColumns: Array[String] = Array.empty`
- `bucketSpec: Option[BucketSpec] = None`
- `provider: String`
- `options: Map[String, String]`
This class is responsible for deciding which of the Data Source APIs a given provider is using (including the non-file based ones). All reconciliation of partitions, buckets, schema from metastores or inference is done here.
### DataSourceAnalysis / DataSourceStrategy
Responsible for analyzing and planning reading/writing of data using any of the Data Source APIs, including:
- pruning the files from partitions that will be read based on filters.
- appending partition columns*
- applying additional filters when a data source can not evaluate them internally.
- constructing an RDD that is bucketed correctly when required*
- sanity checking schema match-up and other analysis when writing.
*In the future we should do that following:
- Break out file handling into its own Strategy as its sufficiently complex / isolated.
- Push the appending of partition columns down in to `FileFormat` to avoid an extra copy / unvectorization.
- Use a custom RDD for scans instead of `SQLNewNewHadoopRDD2`
Author: Michael Armbrust <michael@databricks.com>
Author: Wenchen Fan <wenchen@databricks.com>
Closes#11509 from marmbrus/fileDataSource.
Add save/load for feature.py. Meanwhile, add save/load for `ElementwiseProduct` in Scala side and fix a bug of missing `setDefault` in `VectorSlicer` and `StopWordsRemover`.
In this PR I ignore the `RFormula` and `RFormulaModel` because its Scala implementation is pending in https://github.com/apache/spark/pull/9884. I'll add them in this PR if https://github.com/apache/spark/pull/9884 gets merged first. Or add a follow-up JIRA for `RFormula`.
Author: Xusen Yin <yinxusen@gmail.com>
Closes#11203 from yinxusen/SPARK-13036.
## What changes were proposed in this pull request?
It avoids counting the dataframe twice.
Author: Abou Haydar Elias <abouhaydar.elias@gmail.com>
Author: Elie A <abouhaydar.elias@gmail.com>
Closes#11491 from eliasah/quantile-discretizer-patch.
## What changes were proposed in this pull request?
This PR fixes typos in comments and testcase name of code.
## How was this patch tested?
manual.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11481 from dongjoon-hyun/minor_fix_typos_in_code.
## What changes were proposed in this pull request?
Remove duplicated periods at the end of some sharedParams in ScalaDoc, such as [here](https://github.com/apache/spark/pull/11344/files#diff-9edc669edcf2c0c7cf1efe4a0a57da80L367)
cc mengxr srowen
## How was this patch tested?
Documents change, no test.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11344 from yanboliang/shared-cleanup.
## What changes were proposed in this pull request?
After SPARK-6990, `dev/lint-java` keeps Java code healthy and helps PR review by saving much time.
This issue aims remove unused imports from Java/Scala code and add `UnusedImports` checkstyle rule to help developers.
## How was this patch tested?
```
./dev/lint-java
./build/sbt compile
```
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11438 from dongjoon-hyun/SPARK-13583.
## What changes were proposed in this pull request?
Make some cross-cutting code improvements according to static analysis. These are individually up for discussion since they exist in separate commits that can be reverted. The changes are broadly:
- Inner class should be static
- Mismatched hashCode/equals
- Overflow in compareTo
- Unchecked warnings
- Misuse of assert, vs junit.assert
- get(a) + getOrElse(b) -> getOrElse(a,b)
- Array/String .size -> .length (occasionally, -> .isEmpty / .nonEmpty) to avoid implicit conversions
- Dead code
- tailrec
- exists(_ == ) -> contains find + nonEmpty -> exists filter + size -> count
- reduce(_+_) -> sum map + flatten -> map
The most controversial may be .size -> .length simply because of its size. It is intended to avoid implicits that might be expensive in some places.
## How was the this patch tested?
Existing Jenkins unit tests.
Author: Sean Owen <sowen@cloudera.com>
Closes#11292 from srowen/SPARK-13423.
Estimator for Generalized Linear Models(GLMs) which will be solved by IRLS.
cc mengxr
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11136 from yanboliang/spark-12811.
JIRA: https://issues.apache.org/jira/browse/SPARK-13506
## What changes were proposed in this pull request?
just chang R Snippet Comment in AssociationRulesSuite
## How was this patch tested?
unit test passsed
Author: Zheng RuiFeng <ruifengz@foxmail.com>
Closes#11387 from zhengruifeng/ars.
## What changes were proposed in this pull request?
* The default value of ```regParam``` of PySpark MLlib ```LogisticRegressionWithLBFGS``` should be consistent with Scala which is ```0.0```. (This is also consistent with ML ```LogisticRegression```.)
* BTW, if we use a known updater(L1 or L2) for binary classification, ```LogisticRegressionWithLBFGS``` will call the ML implementation. We should update the API doc to clarifying ```numCorrections``` will have no effect if we fall into that route.
* Make a pass for all parameters of ```LogisticRegressionWithLBFGS```, others are set properly.
cc mengxr dbtsai
## How was this patch tested?
No new tests, it should pass all current tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11424 from yanboliang/spark-13545.
Part of task for [SPARK-11219](https://issues.apache.org/jira/browse/SPARK-11219) to make PySpark MLlib parameter description formatting consistent. This is for the tree module.
closes#10601
Author: Bryan Cutler <cutlerb@gmail.com>
Author: vijaykiran <mail@vijaykiran.com>
Closes#11353 from BryanCutler/param-desc-consistent-tree-SPARK-12634.
## What changes were proposed in this pull request?
This is another try of PR #11323.
This PR removes DataFrame RDD operations except for `foreach` and `foreachPartitions` (they are actions rather than transformations). Original calls are now replaced by calls to methods of `DataFrame.rdd`.
PR #11323 was reverted because it introduced a regression: both `DataFrame.foreach` and `DataFrame.foreachPartitions` wrap underlying RDD operations with `withNewExecutionId` to track Spark jobs. But they are removed in #11323.
## How was the this patch tested?
No extra tests are added. Existing tests should do the work.
Author: Cheng Lian <lian@databricks.com>
Closes#11388 from liancheng/remove-df-rdd-ops.
jira: https://issues.apache.org/jira/browse/SPARK-13028
MaxAbsScaler works in a very similar way as MinMaxScaler, but scales in a way that the training data lies within the range [-1, 1] by dividing through the largest maximum value in each feature. The motivation to use this scaling includes robustness to very small standard deviations of features and preserving zero entries in sparse data.
Unlike StandardScaler and MinMaxScaler, MaxAbsScaler does not shift/center the data, and thus does not destroy any sparsity.
Something similar from sklearn:
http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MaxAbsScaler.html#sklearn.preprocessing.MaxAbsScaler
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#10939 from hhbyyh/maxabs and squashes the following commits:
fd8bdcd [Yuhao Yang] add tag and some optimization on fit
648fced [Yuhao Yang] Merge remote-tracking branch 'upstream/master' into maxabs
75bebc2 [Yuhao Yang] Merge remote-tracking branch 'upstream/master' into maxabs
cb10bb6 [Yuhao Yang] remove minmax
91ef8f3 [Yuhao Yang] ut added
8ab0747 [Yuhao Yang] Merge remote-tracking branch 'upstream/master' into maxabs
a9215b5 [Yuhao Yang] max abs scaler
## What changes were proposed in this pull request?
ML StringIndexer does not protect itself from column name duplication.
We should still improve a way to validate a schema of `StringIndexer` and `StringIndexerModel`. However, it would be great to fix at another issue.
## How was this patch tested?
unit test
Author: Yu ISHIKAWA <yuu.ishikawa@gmail.com>
Closes#11370 from yu-iskw/SPARK-12874.
## What changes were proposed in this pull request?
This PR removes DataFrame RDD operations. Original calls are now replaced by calls to methods of `DataFrame.rdd`.
## How was the this patch tested?
No extra tests are added. Existing tests should do the work.
Author: Cheng Lian <lian@databricks.com>
Closes#11323 from liancheng/remove-df-rdd-ops.
## What changes were proposed in this pull request?
Like #11027 for ```LogisticRegression```, ```LinearRegression``` with L1 regularization should also cache the value of the ```standardization``` rather than re-fetching it from the ```ParamMap``` for every OWLQN iteration.
cc srowen
## How was this patch tested?
No extra tests are added. It should pass all existing tests.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11367 from yanboliang/spark-13490.
## What changes were proposed in this pull request?
Change line 113 of QuantileDiscretizer.scala to
`val requiredSamples = math.max(numBins * numBins, 10000.0)`
so that `requiredSamples` is a `Double`. This will fix the division in line 114 which currently results in zero if `requiredSamples < dataset.count`
## How was the this patch tested?
Manual tests. I was having a problems using QuantileDiscretizer with my a dataset and after making this change QuantileDiscretizer behaves as expected.
Author: Oliver Pierson <ocp@gatech.edu>
Author: Oliver Pierson <opierson@umd.edu>
Closes#11319 from oliverpierson/SPARK-13444.
`GraphImpl.fromExistingRDDs` expects preprocessed vertex RDD as input. We call it in LDA without validating this requirement. So it might introduce errors. Replacing it by `Graph.apply` would be safer and more proper because it is a public API. The tests still pass. So maybe it is safe to use `fromExistingRDDs` here (though it doesn't seem so based on the implementation) or the test cases are special. jkbradley ankurdave
Author: Xiangrui Meng <meng@databricks.com>
Closes#11226 from mengxr/SPARK-13355.
## What changes were proposed in this pull request?
In order to provide better and consistent result, let's change the default value of MLlib ```LogisticRegressionWithLBFGS convergenceTol``` from ```1E-4``` to ```1E-6``` which will be equal to ML ```LogisticRegression```.
cc dbtsai
## How was the this patch tested?
unit tests
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11299 from yanboliang/spark-13429.
As also mentioned/marked by TODO in AFTAggregator.AFTAggregator.add(data: AFTPoint) method a new array is being created for intercept value and it is being concatenated
with another array which contains the betas, the resulted Array is being converted into a Dense vector which in its turn is being converted into breeze vector.
This is expensive and not necessarily beautiful.
I've tried to solve above mentioned problem by simple algebraic decompositions - keeping and treating intercept independently.
Please let me know what do you think and if you have any questions.
Thanks,
Narine
Author: Narine Kokhlikyan <narine.kokhlikyan@gmail.com>
Closes#11179 from NarineK/survivaloptim.
ML ```KMeansModel / BisectingKMeansModel / QuantileDiscretizer``` should set parent.
cc mengxr
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11214 from yanboliang/spark-13334.
Part of task for [SPARK-11219](https://issues.apache.org/jira/browse/SPARK-11219) to make PySpark MLlib parameter description formatting consistent. This is for the fpm and recommendation modules.
Closes#10602Closes#10897
Author: Bryan Cutler <cutlerb@gmail.com>
Author: somideshmukh <somilde@us.ibm.com>
Closes#11186 from BryanCutler/param-desc-consistent-fpmrecc-SPARK-12632.
## What changes were proposed in this pull request?
This PR tries to fix all typos in all markdown files under `docs` module,
and fixes similar typos in other comments, too.
## How was the this patch tested?
manual tests.
Author: Dongjoon Hyun <dongjoon@apache.org>
Closes#11300 from dongjoon-hyun/minor_fix_typos.
add support of arbitrary length sentence by using the nature representation of sentences in the input.
add new similarity functions and add normalization option for distances in synonym finding
add new accessor for internal structure(the vocabulary and wordindex) for convenience
need instructions about how to set value for the Since annotation for newly added public functions. 1.5.3?
jira link: https://issues.apache.org/jira/browse/SPARK-12153
Author: Yong Gang Cao <ygcao@amazon.com>
Author: Yong-Gang Cao <ygcao@users.noreply.github.com>
Closes#10152 from ygcao/improvementForSentenceBoundary.
## What changes were proposed in this pull request?
Fix MLlib LogisticRegressionWithLBFGS regularization map as:
```SquaredL2Updater``` -> ```elasticNetParam = 0.0```
```L1Updater``` -> ```elasticNetParam = 1.0```
cc dbtsai
## How was the this patch tested?
unit tests
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#11258 from yanboliang/spark-13379.
This PR fixes some warnings found by `build/sbt mllib/test:compile`.
Author: Xiangrui Meng <meng@databricks.com>
Closes#11227 from mengxr/fix-mllib-warnings-201602.
This documents the implementation of ALS in `spark.ml` with example code in scala, java and python.
Author: BenFradet <benjamin.fradet@gmail.com>
Closes#10411 from BenFradet/SPARK-12247.
This enhancement extends the existing SparkML Binarizer [SPARK-5891] to allow Vector in addition to the existing Double input column type.
A use case for this enhancement is for when a user wants to Binarize many similar feature columns at once using the same threshold value (for example a binary threshold applied to many pixels in an image).
This contribution is my original work and I license the work to the project under the project's open source license.
viirya mengxr
Author: seddonm1 <seddonm1@gmail.com>
Closes#10976 from seddonm1/master.
JIRA: https://issues.apache.org/jira/browse/SPARK-12363
This issue is pointed by yanboliang. When `setRuns` is removed from PowerIterationClustering, one of the tests will be failed. I found that some `dstAttr`s of the normalized graph are not correct values but 0.0. By setting `TripletFields.All` in `mapTriplets` it can work.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Author: Xiangrui Meng <meng@databricks.com>
Closes#10539 from viirya/fix-poweriter.
jkbradley I tried to improve the function to export a model. When I tried to export a model to S3 under Spark 1.6, we couldn't do that. So, it should offer S3 besides HDFS. Can you review it when you have time? Thanks!
Author: Yu ISHIKAWA <yuu.ishikawa@gmail.com>
Closes#11151 from yu-iskw/SPARK-13265.
In spark-env.sh.template, there are multi-byte characters, this PR will remove it.
Author: Sasaki Toru <sasakitoa@nttdata.co.jp>
Closes#11149 from sasakitoa/remove_multibyte_in_sparkenv.
JIRA: https://issues.apache.org/jira/browse/SPARK-10524
Currently we use the hard prediction (`ImpurityCalculator.predict`) to order categories' bins. But we should use the soft prediction.
Author: Liang-Chi Hsieh <viirya@gmail.com>
Author: Liang-Chi Hsieh <viirya@appier.com>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8734 from viirya/dt-soft-centroids.
KMeans:
Make a private non-deprecated version of setRuns API so that we can call it from the PythonAPI without deprecation warnings in our own build. Also use it internally when being called from train. Add a logWarning for non-1 values
MFDataGenerator:
Apparently we are calling round on an integer which now in Scala 2.11 results in a warning (it didn't make any sense before either). Figure out if this is a mistake we can just remove or if we got the types wrong somewhere.
I put these two together since they are both deprecation fixes in MLlib and pretty small, but I can split them up if we would prefer it that way.
Author: Holden Karau <holden@us.ibm.com>
Closes#11112 from holdenk/SPARK-13201-non-deprecated-setRuns-SPARK-mathround-integer.
cache the value of the standardization Param in LogisticRegression, rather than re-fetching it from the ParamMap for every index and every optimization step in the quasi-newton optimizer
also, fix Param#toString to cache the stringified representation, rather than re-interpolating it on every call, so any other implementations that have similar repeated access patterns will see a benefit.
this change improves training times for one of my test sets from ~7m30s to ~4m30s
Author: Gary King <gary@idibon.com>
Closes#11027 from idigary/spark-13132-optimize-logistic-regression.
Fixed the bug in linear regression train for the case when the target variable is constant. The two cases for `fitIntercept=true` or `fitIntercept=false` should be treated differently.
Author: Imran Younus <iyounus@us.ibm.com>
Closes#10702 from iyounus/SPARK-12732_bug_fix_in_linear_regression_train.
Fixes problem and verifies fix by test suite.
Also - adds optional parameter: nullable (Boolean) to: SchemaUtils.appendColumn
and deduplicates SchemaUtils.appendColumn functions.
Author: Grzegorz Chilkiewicz <grzegorz.chilkiewicz@codilime.com>
Closes#10741 from grzegorz-chilkiewicz/master.
Part of task for [SPARK-11219](https://issues.apache.org/jira/browse/SPARK-11219) to make PySpark MLlib parameter description formatting consistent. This is for the clustering module.
Author: Bryan Cutler <cutlerb@gmail.com>
Closes#10610 from BryanCutler/param-desc-consistent-cluster-SPARK-12631.
This patch changes Spark's build to make Scala 2.11 the default Scala version. To be clear, this does not mean that Spark will stop supporting Scala 2.10: users will still be able to compile Spark for Scala 2.10 by following the instructions on the "Building Spark" page; however, it does mean that Scala 2.11 will be the default Scala version used by our CI builds (including pull request builds).
The Scala 2.11 compiler is faster than 2.10, so I think we'll be able to look forward to a slight speedup in our CI builds (it looks like it's about 2X faster for the Maven compile-only builds, for instance).
After this patch is merged, I'll update Jenkins to add new compile-only jobs to ensure that Scala 2.10 compilation doesn't break.
Author: Josh Rosen <joshrosen@databricks.com>
Closes#10608 from JoshRosen/SPARK-6363.
Implement ```IterativelyReweightedLeastSquares``` solver for GLM. I consider it as a solver rather than estimator, it only used internal so I keep it ```private[ml]```.
There are two limitations in the current implementation compared with R:
* It can not support ```Tuple``` as response for ```Binomial``` family, such as the following code:
```
glm( cbind(using, notUsing) ~ age + education + wantsMore , family = binomial)
```
* It does not support ```offset```.
Because I considered that ```RFormula``` did not support ```Tuple``` as label and ```offset``` keyword, so I simplified the implementation. But to add support for these two functions is not very hard, I can do it in follow-up PR if it is necessary. Meanwhile, we can also add R-like statistic summary for IRLS.
The implementation refers R, [statsmodels](https://github.com/statsmodels/statsmodels) and [sparkGLM](https://github.com/AlteryxLabs/sparkGLM).
Please focus on the main structure and overpass minor issues/docs that I will update later. Any comments and opinions will be appreciated.
cc mengxr jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#10639 from yanboliang/spark-9835.
The intercept in Logistic Regression represents a prior on categories which should not be regularized. In MLlib, the regularization is handled through Updater, and the Updater penalizes all the components without excluding the intercept which resulting poor training accuracy with regularization.
The new implementation in ML framework handles this properly, and we should call the implementation in ML from MLlib since majority of users are still using MLlib api.
Note that both of them are doing feature scalings to improve the convergence, and the only difference is ML version doesn't regularize the intercept. As a result, when lambda is zero, they will converge to the same solution.
Previously partially reviewed at https://github.com/apache/spark/pull/6386#issuecomment-168781424 re-opening for dbtsai to review.
Author: Holden Karau <holden@us.ibm.com>
Author: Holden Karau <holden@pigscanfly.ca>
Closes#10788 from holdenk/SPARK-7780-intercept-in-logisticregressionwithLBFGS-should-not-be-regularized.
… Add LibSVMOutputWriter
The behavior of LibSVMRelation is not changed except adding LibSVMOutputWriter
* Partition is still not supported
* Multiple input paths is not supported
Author: Jeff Zhang <zjffdu@apache.org>
Closes#9595 from zjffdu/SPARK-11622.
https://issues.apache.org/jira/browse/SPARK-12834
We use `SerDe.dumps()` to serialize `JavaArray` and `JavaList` in `PythonMLLibAPI`, then deserialize them with `PickleSerializer` in Python side. However, there is no need to transform them in such an inefficient way. Instead of it, we can use type conversion to convert them, e.g. `list(JavaArray)` or `list(JavaList)`. What's more, there is an issue to Ser/De Scala Array as I said in https://issues.apache.org/jira/browse/SPARK-12780
Author: Xusen Yin <yinxusen@gmail.com>
Closes#10772 from yinxusen/SPARK-12834.
```PCAModel``` can output ```explainedVariance``` at Python side.
cc mengxr srowen
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#10830 from yanboliang/spark-12905.
Update user guide for RFormula feature interactions. Meanwhile we also update other new features such as supporting string label in Spark 1.6.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#10222 from yanboliang/spark-11965.
- Remove Akka dependency from core. Note: the streaming-akka project still uses Akka.
- Remove HttpFileServer
- Remove Akka configs from SparkConf and SSLOptions
- Rename `spark.akka.frameSize` to `spark.rpc.message.maxSize`. I think it's still worth to keep this config because using `DirectTaskResult` or `IndirectTaskResult` depends on it.
- Update comments and docs
Author: Shixiong Zhu <shixiong@databricks.com>
Closes#10854 from zsxwing/remove-akka.
When all labels are the same, it's a dangerous ground for LogisticRegression without intercept to converge. GLMNET doesn't support this case, and will just exit. GLM can train, but will have a warning message saying the algorithm doesn't converge.
Author: DB Tsai <dbt@netflix.com>
Closes#10862 from dbtsai/add-tests.
Add Since annotations to ml.param and ml.*
Author: Takahashi Hiroshi <takahashi.hiroshi@lab.ntt.co.jp>
Author: Hiroshi Takahashi <takahashi.hiroshi@lab.ntt.co.jp>
Closes#8935 from taishi-oss/issue10263.
This fixes the behavior of WeightedLeastSquars.fit() when the standard deviation of the target variable is zero. If the fitIntercept is true, there is no need to train.
Author: Imran Younus <iyounus@us.ibm.com>
Closes#10274 from iyounus/SPARK-12230_bug_fix_in_weighted_least_squares.
This PR aims to allow the prediction column of `BinaryClassificationEvaluator` to be of double type.
Author: BenFradet <benjamin.fradet@gmail.com>
Closes#10472 from BenFradet/SPARK-9716.
From the coverage issues for 1.6 : Add Python API for mllib.clustering.BisectingKMeans.
Author: Holden Karau <holden@us.ibm.com>
Closes#10150 from holdenk/SPARK-11937-python-api-coverage-SPARK-11944-python-mllib.clustering.BisectingKMeans.
Change assertion's message so it's consistent with the code. The old message says that the invoked method was lapack.dports, where in fact it was lapack.dppsv method.
Author: Wojciech Jurczyk <wojtek.jurczyk@gmail.com>
Closes#10818 from wjur/wjur/rename_error_message.
Currently `summary()` fails on a GLM model fitted over a vector feature missing ML attrs, since the output feature attrs will also have no name. We can avoid this situation by forcing `VectorAssembler` to make up suitable names when inputs are missing names.
cc mengxr
Author: Eric Liang <ekl@databricks.com>
Closes#10323 from ericl/spark-12346.
I create new pr since original pr long time no update.
Please help to review.
srowen
Author: Tommy YU <tummyyu@163.com>
Closes#10756 from Wenpei/add_since_to_recomm.
jira: https://issues.apache.org/jira/browse/SPARK-12026
The issue is valid as features.toArray.view.zipWithIndex.slice(startCol, endCol) becomes slower as startCol gets larger.
I tested on local and the change can improve the performance and the running time was stable.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#10146 from hhbyyh/chiSq.
jira: https://issues.apache.org/jira/browse/SPARK-10809
We could provide a single-document topicDistributions method for LocalLDAModel to allow for quick queries which avoid RDD operations. Currently, the user must use an RDD of documents.
add some missing assert too.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#9484 from hhbyyh/ldaTopicPre.
jira: https://issues.apache.org/jira/browse/SPARK-12685
the log of `word2vec` reports
trainWordsCount = -785727483
during computation over a large dataset.
Update the priority as it will affect the computation process.
`alpha = learningRate * (1 - numPartitions * wordCount.toDouble / (trainWordsCount + 1))`
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#10627 from hhbyyh/w2voverflow.
PySpark MLlib ```GaussianMixtureModel``` should support single instance ```predict/predictSoft``` just like Scala do.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#10552 from yanboliang/spark-12603.
Turn import ordering violations into build errors, plus a few adjustments
to account for how the checker behaves. I'm a little on the fence about
whether the existing code is right, but it's easier to appease the checker
than to discuss what's the more correct order here.
Plus a few fixes to imports that cropped in since my recent cleanups.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#10612 from vanzin/SPARK-3873-enable.
Fix the style violation (space before , and :).
This PR is a followup for #10643.
Author: Kousuke Saruta <sarutak@oss.nttdata.co.jp>
Closes#10684 from sarutak/SPARK-12692-followup-mllib.
Fix most build warnings: mostly deprecated API usages. I'll annotate some of the changes below. CC rxin who is leading the charge to remove the deprecated APIs.
Author: Sean Owen <sowen@cloudera.com>
Closes#10570 from srowen/SPARK-12618.
For the BinaryClassificationEvaluator, the scaladoc doesn't mention that "areaUnderPR" is supported, only that the default is "areadUnderROC".
Also, in the documentation, it is said that:
"The default metric used to choose the best ParamMap can be overriden by the setMetric method in each of these evaluators."
However, the method is called setMetricName.
This PR aims to fix both issues.
Author: BenFradet <benjamin.fradet@gmail.com>
Closes#10328 from BenFradet/SPARK-12368.
Modified the definition of R^2 for regression through origin. Added modified test for regression metrics.
Author: Imran Younus <iyounus@us.ibm.com>
Author: Imran Younus <imranyounus@gmail.com>
Closes#10384 from iyounus/SPARK_12331_R2_for_regression_through_origin.
DecisionTreeRegressor will provide variance of prediction as a Double column.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8866 from yanboliang/spark-9622.
callUDF has been deprecated. However, we do not have an alternative for users to specify the output data type without type tags. This pull request introduced a new API for that, and replaces the invocation of the deprecated callUDF with that.
Author: Reynold Xin <rxin@databricks.com>
Closes#10547 from rxin/SPARK-12599.
A slight adjustment to the checker configuration was needed; there is
a handful of warnings still left, but those are because of a bug in
the checker that I'll fix separately (before enabling errors for the
checker, of course).
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#10535 from vanzin/SPARK-3873-mllib.
Include the following changes:
1. Close `java.sql.Statement`
2. Fix incorrect `asInstanceOf`.
3. Remove unnecessary `synchronized` and `ReentrantLock`.
Author: Shixiong Zhu <shixiong@databricks.com>
Closes#10440 from zsxwing/findbugs.
ParamMap#filter uses `mutable.Map#filterKeys`. The return type of `filterKey` is collection.Map, not mutable.Map but the result is casted to mutable.Map using `asInstanceOf` so we get `ClassCastException`.
Also, the return type of Map#filterKeys is not Serializable. It's the issue of Scala (https://issues.scala-lang.org/browse/SI-6654).
Author: Kousuke Saruta <sarutak@oss.nttdata.co.jp>
Closes#10381 from sarutak/SPARK-12424.
Restore the original value of os.arch property after each test
Since some of tests forced to set the specific value to os.arch property, we need to set the original value.
Author: Kazuaki Ishizaki <ishizaki@jp.ibm.com>
Closes#10289 from kiszk/SPARK-12311.
Only load explainedVariance in PCAModel if it was written with Spark > 1.6.x
jkbradley is this kind of what you had in mind?
Author: Sean Owen <sowen@cloudera.com>
Closes#10327 from srowen/SPARK-12349.
Added catch for casting Long to Int exception when PySpark ALS Ratings are serialized. It is easy to accidentally use Long IDs for user/product and before, it would fail with a somewhat cryptic "ClassCastException: java.lang.Long cannot be cast to java.lang.Integer." Now if this is done, a more descriptive error is shown, e.g. "PickleException: Ratings id 1205640308657491975 exceeds max integer value of 2147483647."
Author: Bryan Cutler <bjcutler@us.ibm.com>
Closes#9361 from BryanCutler/als-pyspark-long-id-error-SPARK-10158.
Use ```sqlContext``` from ```MLlibTestSparkContext``` rather than creating new one for spark.ml test suites. I have checked thoroughly and found there are four test cases need to update.
cc mengxr jkbradley
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#10279 from yanboliang/spark-12309.
JIRA: https://issues.apache.org/jira/browse/SPARK-12016
We should not directly use Word2VecModel in pyspark. We need to wrap it in a Word2VecModelWrapper when loading it in pyspark.
Author: Liang-Chi Hsieh <viirya@appier.com>
Closes#10100 from viirya/fix-load-py-wordvecmodel.
As noted in PR #9441, implementing `tallSkinnyQR` uncovered a bug with our PySpark `RowMatrix` constructor. As discussed on the dev list [here](http://apache-spark-developers-list.1001551.n3.nabble.com/K-Means-And-Class-Tags-td10038.html), there appears to be an issue with type erasure with RDDs coming from Java, and by extension from PySpark. Although we are attempting to construct a `RowMatrix` from an `RDD[Vector]` in [PythonMLlibAPI](https://github.com/apache/spark/blob/master/mllib/src/main/scala/org/apache/spark/mllib/api/python/PythonMLLibAPI.scala#L1115), the `Vector` type is erased, resulting in an `RDD[Object]`. Thus, when calling Scala's `tallSkinnyQR` from PySpark, we get a Java `ClassCastException` in which an `Object` cannot be cast to a Spark `Vector`. As noted in the aforementioned dev list thread, this issue was also encountered with `DecisionTrees`, and the fix involved an explicit `retag` of the RDD with a `Vector` type. `IndexedRowMatrix` and `CoordinateMatrix` do not appear to have this issue likely due to their related helper functions in `PythonMLlibAPI` creating the RDDs explicitly from DataFrames with pattern matching, thus preserving the types.
This PR currently contains that retagging fix applied to the `createRowMatrix` helper function in `PythonMLlibAPI`. This PR blocks #9441, so once this is merged, the other can be rebased.
cc holdenk
Author: Mike Dusenberry <mwdusenb@us.ibm.com>
Closes#9458 from dusenberrymw/SPARK-11497_PySpark_RowMatrix_Constructor_Has_Type_Erasure_Issue.
LogisticRegression training summary should still function if the predictionCol is set to an empty string or otherwise unset (related too https://issues.apache.org/jira/browse/SPARK-9718 )
Author: Holden Karau <holden@pigscanfly.ca>
Author: Holden Karau <holden@us.ibm.com>
Closes#9037 from holdenk/SPARK-10991-LogisticRegressionTrainingSummary-handle-empty-prediction-col.
Add `computePrincipalComponentsAndVariance` to also compute PCA's explained variance.
CC mengxr
Author: Sean Owen <sowen@cloudera.com>
Closes#9736 from srowen/SPARK-11530.
Currently word2vec has the window hard coded at 5, some users may want different sizes (for example if using on n-gram input or similar). User request comes from http://stackoverflow.com/questions/32231975/spark-word2vec-window-size .
Author: Holden Karau <holden@us.ibm.com>
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8513 from holdenk/SPARK-10299-word2vec-should-allow-users-to-specify-the-window-size.
felixcheung , mengxr
Just added a message to require()
Author: Dominik Dahlem <dominik.dahlem@gmail.combination>
Closes#9598 from dahlem/ddahlem_regression_evaluator_double_predictions_message_04112015.
jira: https://issues.apache.org/jira/browse/SPARK-11605
Check Java compatibility for MLlib for this release.
fix:
1. `StreamingTest.registerStream` needs java friendly interface.
2. `GradientBoostedTreesModel.computeInitialPredictionAndError` and `GradientBoostedTreesModel.updatePredictionError` has java compatibility issue. Mark them as `developerAPI`.
TBD:
[updated] no fix for now per discussion.
`org.apache.spark.mllib.classification.LogisticRegressionModel`
`public scala.Option<java.lang.Object> getThreshold();` has wrong return type for Java invocation.
`SVMModel` has the similar issue.
Yet adding a `scala.Option<java.util.Double> getThreshold()` would result in an overloading error due to the same function signature. And adding a new function with different name seems to be not necessary.
cc jkbradley feynmanliang
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#10102 from hhbyyh/javaAPI.
Sparse feature generated in LinearDataGenerator does not create dense vectors as an intermediate any more.
Author: Nakul Jindal <njindal@us.ibm.com>
Closes#9756 from nakul02/SPARK-11439_sparse_without_creating_dense_feature.
Add ```SQLTransformer``` user guide, example code and make Scala API doc more clear.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#10006 from yanboliang/spark-11958.
Switched from using SQLContext constructor to using getOrCreate, mainly in model save/load methods.
This covers all instances in spark.mllib. There were no uses of the constructor in spark.ml.
CC: mengxr yhuai
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#10161 from jkbradley/mllib-sqlcontext-fix.
jira: https://issues.apache.org/jira/browse/SPARK-12096
word2vec now can handle much bigger vocabulary.
The old constraint vocabSize.toLong * vectorSize < Ine.max / 8 should be removed.
new constraint is vocabSize.toLong * vectorSize < max array length (usually a little less than Int.MaxValue)
I tested with vocabsize over 18M and vectorsize = 100.
srowen jkbradley Sorry to miss this in last PR. I was reminded today.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#10103 from hhbyyh/w2vCapacity.
We should upgrade to SBT 0.13.9, since this is a requirement in order to use SBT's new Maven-style resolution features (which will be done in a separate patch, because it's blocked by some binary compatibility issues in the POM reader plugin).
I also upgraded Scalastyle to version 0.8.0, which was necessary in order to fix a Scala 2.10.5 compatibility issue (see https://github.com/scalastyle/scalastyle/issues/156). The newer Scalastyle is slightly stricter about whitespace surrounding tokens, so I fixed the new style violations.
Author: Josh Rosen <joshrosen@databricks.com>
Closes#10112 from JoshRosen/upgrade-to-sbt-0.13.9.
This replaces https://github.com/apache/spark/pull/9696
Invoke Checkstyle and print any errors to the console, failing the step.
Use Google's style rules modified according to
https://cwiki.apache.org/confluence/display/SPARK/Spark+Code+Style+Guide
Some important checks are disabled (see TODOs in `checkstyle.xml`) due to
multiple violations being present in the codebase.
Suggest fixing those TODOs in a separate PR(s).
More on Checkstyle can be found on the [official website](http://checkstyle.sourceforge.net/).
Sample output (from [build 46345](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/46345/consoleFull)) (duplicated because I run the build twice with different profiles):
> Checkstyle checks failed at following occurrences:
[ERROR] src/main/java/org/apache/spark/sql/execution/datasources/parquet/UnsafeRowParquetRecordReader.java:[217,7] (coding) MissingSwitchDefault: switch without "default" clause.
> [ERROR] src/main/java/org/apache/spark/sql/execution/datasources/parquet/SpecificParquetRecordReaderBase.java:[198,10] (modifier) ModifierOrder: 'protected' modifier out of order with the JLS suggestions.
> [ERROR] src/main/java/org/apache/spark/sql/execution/datasources/parquet/UnsafeRowParquetRecordReader.java:[217,7] (coding) MissingSwitchDefault: switch without "default" clause.
> [ERROR] src/main/java/org/apache/spark/sql/execution/datasources/parquet/SpecificParquetRecordReaderBase.java:[198,10] (modifier) ModifierOrder: 'protected' modifier out of order with the JLS suggestions.
> [error] running /home/jenkins/workspace/SparkPullRequestBuilder2/dev/lint-java ; received return code 1
Also fix some of the minor violations that didn't require sweeping changes.
Apologies for the previous botched PRs - I finally figured out the issue.
cr: JoshRosen, pwendell
> I state that the contribution is my original work, and I license the work to the project under the project's open source license.
Author: Dmitry Erastov <derastov@gmail.com>
Closes#9867 from dskrvk/master.
This fixes SPARK-12000, verified on my local with JDK 7. It seems that `scaladoc` try to match method names and messed up with annotations.
cc: JoshRosen jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#10114 from mengxr/SPARK-12000.2.
cc mengxr noel-smith
I worked on this issues based on https://github.com/apache/spark/pull/8729.
ehsanmok thank you for your contricution!
Author: Yu ISHIKAWA <yuu.ishikawa@gmail.com>
Author: Ehsan M.Kermani <ehsanmo1367@gmail.com>
Closes#9338 from yu-iskw/JIRA-10266.
jira: https://issues.apache.org/jira/browse/SPARK-11898
syn0Global and sync1Global in word2vec are quite large objects with size (vocab * vectorSize * 8), yet they are passed to worker using basic task serialization.
Use broadcast can greatly improve the performance. My benchmark shows that, for 1M vocabulary and default vectorSize 100, changing to broadcast can help,
1. decrease the worker memory consumption by 45%.
2. decrease running time by 40%.
This will also help extend the upper limit for Word2Vec.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#9878 from hhbyyh/w2vBC.
Add read/write support to LDA, similar to ALS.
save/load for ml.LocalLDAModel is done.
For DistributedLDAModel, I'm not sure if we can invoke save on the mllib.DistributedLDAModel directly. I'll send update after some test.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#9894 from hhbyyh/ldaMLsave.
Doc for 1.6 that the summaries mostly ignore the weight column.
To be corrected for 1.7
CC: mengxr thunterdb
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9927 from jkbradley/linregsummary-doc.
There is an unhandled case in the transform method of VectorAssembler if one of the input columns doesn't have one of the supported type DoubleType, NumericType, BooleanType or VectorUDT.
So, if you try to transform a column of StringType you get a cryptic "scala.MatchError: StringType".
This PR aims to fix this, throwing a SparkException when dealing with an unknown column type.
Author: BenFradet <benjamin.fradet@gmail.com>
Closes#9885 from BenFradet/SPARK-11902.
Like [SPARK-11852](https://issues.apache.org/jira/browse/SPARK-11852), ```k``` is params and we should save it under ```metadata/``` rather than both under ```data/``` and ```metadata/```. Refactor the constructor of ```ml.feature.PCAModel``` to take only ```pc``` but construct ```mllib.feature.PCAModel``` inside ```transform```.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9897 from yanboliang/spark-11912.
I believe this works for general estimators within CrossValidator, including compound estimators. (See the complex unit test.)
Added read/write for all 3 Evaluators as well.
CC: mengxr yanboliang
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9848 from jkbradley/cv-io.
```withStd``` and ```withMean``` should be params of ```StandardScaler``` and ```StandardScalerModel```.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9839 from yanboliang/standardScaler-refactor.
Need to remove parent directory (```className```) rather than just tempDir (```className/random_name```)
I tested this with IDFSuite, which has 2 read/write tests, and it fixes the problem.
CC: mengxr Can you confirm this is fine? I believe it is since the same ```random_name``` is used for all tests in a suite; we basically have an extra unneeded level of nesting.
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9851 from jkbradley/tempdir-cleanup.
Add read/write support to the following estimators under spark.ml:
* ChiSqSelector
* PCA
* VectorIndexer
* Word2Vec
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9838 from yanboliang/spark-11829.
Updates:
* Add repartition(1) to save() methods' saving of data for LogisticRegressionModel, LinearRegressionModel.
* Strengthen privacy to class and companion object for Writers and Readers
* Change LogisticRegressionSuite read/write test to fit intercept
* Add Since versions for read/write methods in Pipeline, LogisticRegression
* Switch from hand-written class names in Readers to using getClass
CC: mengxr
CC: yanboliang Would you mind taking a look at this PR? mengxr might not be able to soon. Thank you!
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9829 from jkbradley/ml-io-cleanups.
* add "ML" prefix to reader/writer/readable/writable to avoid name collision with java.util.*
* define `DefaultParamsReadable/Writable` and use them to save some code
* use `super.load` instead so people can jump directly to the doc of `Readable.load`, which documents the Java compatibility issues
jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#9827 from mengxr/SPARK-11839.
Add read/write support to the following estimators under spark.ml:
* CountVectorizer
* IDF
* MinMaxScaler
* StandardScaler (a little awkward because we store some params in spark.mllib model)
* StringIndexer
Added some necessary method for read/write. Maybe we should add `private[ml] trait DefaultParamsReadable` and `DefaultParamsWritable` to save some boilerplate code, though we still need to override `load` for Java compatibility.
jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#9798 from mengxr/SPARK-6787.
This PR includes:
* Update SparkR:::glm, SparkR:::summary API docs.
* Update SparkR machine learning user guide and example codes to show:
* supporting feature interaction in R formula.
* summary for gaussian GLM model.
* coefficients for binomial GLM model.
mengxr
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9727 from yanboliang/spark-11684.
jira: https://issues.apache.org/jira/browse/SPARK-11813
I found the problem during training a large corpus. Avoid serialization of vocab in Word2Vec has 2 benefits.
1. Performance improvement for less serialization.
2. Increase the capacity of Word2Vec a lot.
Currently in the fit of word2vec, the closure mainly includes serialization of Word2Vec and 2 global table.
the main part of Word2vec is the vocab of size: vocab * 40 * 2 * 4 = 320 vocab
2 global table: vocab * vectorSize * 8. If vectorSize = 20, that's 160 vocab.
Their sum cannot exceed Int.max due to the restriction of ByteArrayOutputStream. In any case, avoiding serialization of vocab helps decrease the size of the closure serialization, especially when vectorSize is small, thus to allow larger vocabulary.
Actually there's another possible fix, make local copy of fields to avoid including Word2Vec in the closure. Let me know if that's preferred.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#9803 from hhbyyh/w2vVocab.
Also modifies DefaultParamsWriter.saveMetadata to take optional extra metadata.
CC: mengxr yanboliang
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9786 from jkbradley/als-io.
This replaces [https://github.com/apache/spark/pull/9656] with updates.
fayeshine should be the main author when this PR is committed.
CC: mengxr fayeshine
Author: Wenjian Huang <nextrush@163.com>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9814 from jkbradley/fayeshine-patch-6790.
I have added unit test for ML's StandardScaler By comparing with R's output, please review for me.
Thx.
Author: RoyGaoVLIS <roygao@zju.edu.cn>
Closes#6665 from RoyGao/7013.
This PR makes the default read/write work with simple transformers/estimators that have params of type `Param[Vector]`. jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#9776 from mengxr/SPARK-11764.
Add save/load to LogisticRegression Estimator, and refactor tests a little to make it easier to add similar support to other Estimator, Model pairs.
Moved LogisticRegressionReader/Writer to within LogisticRegressionModel
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9749 from jkbradley/lr-io-2.
This excludes Estimators and ones which include Vector and other non-basic types for Params or data. This adds:
* Bucketizer
* DCT
* HashingTF
* Interaction
* NGram
* Normalizer
* OneHotEncoder
* PolynomialExpansion
* QuantileDiscretizer
* RFormula
* SQLTransformer
* StopWordsRemover
* StringIndexer
* Tokenizer
* VectorAssembler
* VectorSlicer
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9755 from jkbradley/transformer-io.
This is to support JSON serialization of Param[Vector] in the pipeline API. It could be used for other purposes too. The schema is the same as `VectorUDT`. jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#9751 from mengxr/SPARK-11766.
Pipeline and PipelineModel extend Readable and Writable. Persistence succeeds only when all stages are Writable.
Note: This PR reinstates tests for other read/write functionality. It should probably not get merged until [https://issues.apache.org/jira/browse/SPARK-11672] gets fixed.
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9674 from jkbradley/pipeline-io.
Use LibSVM data source rather than MLUtils.loadLibSVMFile to load DataFrame, include:
* Use libSVM data source for all example codes under examples/ml, and remove unused import.
* Use libSVM data source for user guides under ml-*** which were omitted by #8697.
* Fix bug: We should use ```sqlContext.read().format("libsvm").load(path)``` at Java side, but the API doc and user guides misuse as ```sqlContext.read.format("libsvm").load(path)```.
* Code cleanup.
mengxr
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9690 from yanboliang/spark-11723.
We set `sqlContext = null` in `afterAll`. However, this doesn't change `SQLContext.activeContext` and then `SQLContext.getOrCreate` might use the `SparkContext` from previous test suite and hence causes the error. This PR calls `clearActive` in `beforeAll` and `afterAll` to avoid using an old context from other test suites.
cc: yhuai
Author: Xiangrui Meng <meng@databricks.com>
Closes#9677 from mengxr/SPARK-11672.2.
Per discussion in the initial Pipelines LDA PR [https://github.com/apache/spark/pull/9513], we should make LDAModel abstract and create a LocalLDAModel. This code simplification should be done before the 1.6 release to ensure API compatibility in future releases.
CC feynmanliang mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9678 from jkbradley/lda-pipelines-2.
This causes compile failure with Scala 2.11. See https://issues.scala-lang.org/browse/SI-8813. (Jenkins won't test Scala 2.11. I tested compile locally.) JoshRosen
Author: Xiangrui Meng <meng@databricks.com>
Closes#9644 from mengxr/SPARK-11674.
org.apache.spark.ml.feature.Word2Vec.transform() very slow. we should not read broadcast every sentence.
Author: Yuming Wang <q79969786@gmail.com>
Author: yuming.wang <q79969786@gmail.com>
Author: Xiangrui Meng <meng@databricks.com>
Closes#9592 from 979969786/master.
This PR adds model save/load for spark.ml's LogisticRegressionModel. It also does minor refactoring of the default save/load classes to reuse code.
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9606 from jkbradley/logreg-io2.
This adds LDA to spark.ml, the Pipelines API. It follows the design doc in the JIRA: [https://issues.apache.org/jira/browse/SPARK-5565], with one major change:
* I eliminated doc IDs. These are not necessary with DataFrames since the user can add an ID column as needed.
Note: This will conflict with [https://github.com/apache/spark/pull/9484], but I'll try to merge [https://github.com/apache/spark/pull/9484] first and then rebase this PR.
CC: hhbyyh feynmanliang If you have a chance to make a pass, that'd be really helpful--thanks! Now that I'm done traveling & this PR is almost ready, I'll see about reviewing other PRs critical for 1.6.
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9513 from jkbradley/lda-pipelines.
Implementation of step capability for sliding window function in MLlib's RDD.
Though one can use current sliding window with step 1 and then filter every Nth window, it will take more time and space (N*data.count times more than needed). For example, below are the results for various windows and steps on 10M data points:
Window | Step | Time | Windows produced
------------ | ------------- | ---------- | ----------
128 | 1 | 6.38 | 9999873
128 | 10 | 0.9 | 999988
128 | 100 | 0.41 | 99999
1024 | 1 | 44.67 | 9998977
1024 | 10 | 4.74 | 999898
1024 | 100 | 0.78 | 99990
```
import org.apache.spark.mllib.rdd.RDDFunctions._
val rdd = sc.parallelize(1 to 10000000, 10)
rdd.count
val window = 1024
val step = 1
val t = System.nanoTime(); val windows = rdd.sliding(window, step); println(windows.count); println((System.nanoTime() - t) / 1e9)
```
Author: unknown <ulanov@ULANOV3.americas.hpqcorp.net>
Author: Alexander Ulanov <nashb@yandex.ru>
Author: Xiangrui Meng <meng@databricks.com>
Closes#5855 from avulanov/SPARK-7316-sliding.
Refactoring
* separated overwrite and param save logic in DefaultParamsWriter
* added sparkVersion to DefaultParamsWriter
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#9587 from jkbradley/logreg-io.
jira: https://issues.apache.org/jira/browse/SPARK-11069
quotes from jira:
Tokenizer converts strings to lowercase automatically, but RegexTokenizer does not. It would be nice to add an option to RegexTokenizer to convert to lowercase. Proposal:
call the Boolean Param "toLowercase"
set default to false (so behavior does not change)
Actually sklearn converts to lowercase before tokenizing too
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#9092 from hhbyyh/tokenLower.
I implemented a hierarchical clustering algorithm again. This PR doesn't include examples, documentation and spark.ml APIs. I am going to send another PRs later.
https://issues.apache.org/jira/browse/SPARK-6517
- This implementation based on a bi-sectiong K-means clustering.
- It derives from the freeman-lab 's implementation
- The basic idea is not changed from the previous version. (#2906)
- However, It is 1000x faster than the previous version through parallel processing.
Thank you for your great cooperation, RJ Nowling(rnowling), Jeremy Freeman(freeman-lab), Xiangrui Meng(mengxr) and Sean Owen(srowen).
Author: Yu ISHIKAWA <yuu.ishikawa@gmail.com>
Author: Xiangrui Meng <meng@databricks.com>
Author: Yu ISHIKAWA <yu-iskw@users.noreply.github.com>
Closes#5267 from yu-iskw/new-hierarchical-clustering.
The current pmml models generated do not specify the pmml version in its root node. This is a problem when using this pmml model in other tools because they expect the version attribute to be set explicitly. This fix adds the pmml version attribute to the generated pmml models and specifies its value as 4.2.
Author: fazlan-nazeem <fazlann@wso2.com>
Closes#9558 from fazlan-nazeem/master.
Expose R-like summary statistics in SparkR::glm for linear regression, the output of ```summary``` like
```Java
$DevianceResiduals
Min Max
-0.9509607 0.7291832
$Coefficients
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.6765 0.2353597 7.123139 4.456124e-11
Sepal_Length 0.3498801 0.04630128 7.556598 4.187317e-12
Species_versicolor -0.9833885 0.07207471 -13.64402 0
Species_virginica -1.00751 0.09330565 -10.79796 0
```
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9561 from yanboliang/spark-11494.
Could jkbradley and davies review it?
- Create a wrapper class: `LDAModelWrapper` for `LDAModel`. Because we can't deal with the return value of`describeTopics` in Scala from pyspark directly. `Array[(Array[Int], Array[Double])]` is too complicated to convert it.
- Add `loadLDAModel` in `PythonMLlibAPI`. Since `LDAModel` in Scala is an abstract class and we need to call `load` of `DistributedLDAModel`.
[[SPARK-8467] Add LDAModel.describeTopics() in Python - ASF JIRA](https://issues.apache.org/jira/browse/SPARK-8467)
Author: Yu ISHIKAWA <yuu.ishikawa@gmail.com>
Closes#8643 from yu-iskw/SPARK-8467-2.
This PR implements the default save/load for non-meta estimators and transformers using the JSON serialization of param values. The saved metadata includes:
* class name
* uid
* timestamp
* paramMap
The save/load interface is similar to DataFrames. We use the current active context by default, which should be sufficient for most use cases.
~~~scala
instance.save("path")
instance.write.context(sqlContext).overwrite().save("path")
Instance.load("path")
~~~
The param handling is different from the design doc. We didn't save default and user-set params separately, and when we load it back, all parameters are user-set. This does cause issues. But it also cause other issues if we modify the default params.
TODOs:
* [x] Java test
* [ ] a follow-up PR to implement default save/load for all non-meta estimators and transformers
cc jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#9454 from mengxr/SPARK-11217.
https://issues.apache.org/jira/browse/SPARK-10116
This is really trivial, just happened to notice it -- if `XORShiftRandom.hashSeed` is really supposed to have random bits throughout (as the comment implies), it needs to do something for the conversion to `long`.
mengxr mkolod
Author: Imran Rashid <irashid@cloudera.com>
Closes#8314 from squito/SPARK-10116.
Follow up [SPARK-9836](https://issues.apache.org/jira/browse/SPARK-9836), we should also support summary statistics for ```intercept```.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9485 from yanboliang/spark-11473.
In file LDAOptimizer.scala:
line 441: since "idx" was never used, replaced unrequired zipWithIndex.foreach with foreach.
- nonEmptyDocs.zipWithIndex.foreach { case ((_, termCounts: Vector), idx: Int) =>
+ nonEmptyDocs.foreach { case (_, termCounts: Vector) =>
Author: a1singh <a1singh@ucsd.edu>
Closes#9456 from a1singh/master.
Like ml ```LinearRegression```, ```LogisticRegression``` should provide a training summary including feature names and their coefficients.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9303 from yanboliang/spark-9492.
Currently ```RFormula``` can only handle label with ```NumericType``` or ```BinaryType``` (cast it to ```DoubleType``` as the label of Linear Regression training), we should also support label of ```StringType``` which is needed for Logistic Regression (glm with family = "binomial").
For label of ```StringType```, we should use ```StringIndexer``` to transform it to 0-based index.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9302 from yanboliang/spark-11349.
Removed the old `getModelWeights` function which was private and renamed into `getModelCoefficients`
Author: DB Tsai <dbt@netflix.com>
Closes#9426 from dbtsai/feature-minor.
mengxr, felixcheung
This pull request just relaxes the type of the prediction/label columns to be float and double. Internally, these columns are casted to double. The other evaluators might need to be changed also.
Author: Dominik Dahlem <dominik.dahlem@gmail.combination>
Closes#9296 from dahlem/ddahlem_regression_evaluator_double_predictions_27102015.
This PR deprecates `runs` in k-means. `runs` introduces extra complexity and overhead in MLlib's k-means implementation. I haven't seen much usage with `runs` not equal to `1`. We don't have a unit test for it either. We can deprecate this method in 1.6, and void it in 1.7. It helps us simplify the implementation.
cc: srowen
Author: Xiangrui Meng <meng@databricks.com>
Closes#9322 from mengxr/SPARK-11358.
Made foreachActive public in MLLib's vector API
Author: Nakul Jindal <njindal@us.ibm.com>
Closes#9362 from nakul02/SPARK-11385_foreach_for_mllib_linalg_vector.
…sion as followup. This is the follow up work of SPARK-10668.
* Fix miner style issues.
* Add test case for checking whether solver is selected properly.
Author: Lewuathe <lewuathe@me.com>
Author: lewuathe <lewuathe@me.com>
Closes#9180 from Lewuathe/SPARK-11207.
SparkR glm currently support :
```formula, family = c(“gaussian”, “binomial”), data, lambda = 0, alpha = 0```
We should also support setting standardize which has been defined at [design documentation](https://docs.google.com/document/d/10NZNSEurN2EdWM31uFYsgayIPfCFHiuIu3pCWrUmP_c/edit)
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#9331 from yanboliang/spark-11369.
WeightedLeastSquares now uses the common Instance class in ml.feature instead of a private one.
Author: Nakul Jindal <njindal@us.ibm.com>
Closes#9325 from nakul02/SPARK-11332_refactor_WeightedLeastSquares_dot_Instance.
Fix computation of root-sigma-inverse in multivariate Gaussian; add a test and fix related Python mixture model test.
Supersedes https://github.com/apache/spark/pull/9293
Author: Sean Owen <sowen@cloudera.com>
Closes#9309 from srowen/SPARK-11302.2.
Add columnSimilarities to IndexedRowMatrix by delegating to functionality already in RowMatrix.
With a test.
Author: Reza Zadeh <reza@databricks.com>
Closes#8792 from rezazadeh/colsims.
Remove "Experimental" from .mllib code that has been around since 1.4.0 or earlier
Author: Sean Owen <sowen@cloudera.com>
Closes#9169 from srowen/SPARK-11184.
This is a PR for Parquet-based model import/export.
* Added save/load for ChiSqSelectorModel
* Updated the test suite ChiSqSelectorSuite
Author: Jayant Shekar <jayant@user-MBPMBA-3.local>
Closes#6785 from jayantshekhar/SPARK-6723.
Given row_ind should be less than the number of rows
Given col_ind should be less than the number of cols.
The current code in master gives unpredictable behavior for such cases.
Author: MechCoder <manojkumarsivaraj334@gmail.com>
Closes#8271 from MechCoder/hash_code_matrices.
…2 regularization if the number of features is small
Author: lewuathe <lewuathe@me.com>
Author: Lewuathe <sasaki@treasure-data.com>
Author: Kai Sasaki <sasaki@treasure-data.com>
Author: Lewuathe <lewuathe@me.com>
Closes#8884 from Lewuathe/SPARK-10668.
predictNodeIndex is moved to LearningNode and renamed predictImpl for consistency with Node.predictImpl
Author: Luvsandondov Lkhamsuren <lkhamsurenl@gmail.com>
Closes#8609 from lkhamsurenl/SPARK-9963.
jira: https://issues.apache.org/jira/browse/SPARK-11029
We should add a method analogous to spark.mllib.clustering.KMeansModel.computeCost to spark.ml.clustering.KMeansModel.
This will be a temp fix until we have proper evaluators defined for clustering.
Author: Yuhao Yang <hhbyyh@gmail.com>
Author: yuhaoyang <yuhao@zhanglipings-iMac.local>
Closes#9073 from hhbyyh/computeCost.
This PR aims to decrease communication costs in BlockMatrix multiplication in two ways:
- Simulate the multiplication on the driver, and figure out which blocks actually need to be shuffled
- Send the block once to a partition, and join inside the partition rather than sending multiple copies to the same partition
**NOTE**: One important note is that right now, the old behavior of checking for multiple blocks with the same index is lost. This is not hard to add, but is a little more expensive than how it was.
Initial benchmarking showed promising results (look below), however I did hit some `FileNotFound` exceptions with the new implementation after the shuffle.
Size A: 1e5 x 1e5
Size B: 1e5 x 1e5
Block Sizes: 1024 x 1024
Sparsity: 0.01
Old implementation: 1m 13s
New implementation: 9s
cc avulanov Would you be interested in helping me benchmark this? I used your code from the mailing list (which you sent about 3 months ago?), and the old implementation didn't even run, but the new implementation completed in 268s in a 120 GB / 16 core cluster
Author: Burak Yavuz <brkyvz@gmail.com>
Closes#8757 from brkyvz/opt-bmm.
Value of the quantile probabilities array should be in the range (0, 1) instead of [0,1]
in `AFTSurvivalRegression.scala` according to [Discussion] (https://github.com/apache/spark/pull/8926#discussion-diff-40698242)
Author: vectorijk <jiangkai@gmail.com>
Closes#9083 from vectorijk/spark-11059.
This PR implements the JSON SerDe for the following param types: `Boolean`, `Int`, `Long`, `Float`, `Double`, `String`, `Array[Int]`, `Array[Double]`, and `Array[String]`. The implementation of `Float`, `Double`, and `Array[Double]` are specialized to handle `NaN` and `Inf`s. This will be used in pipeline persistence. jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#9090 from mengxr/SPARK-7402.
Support for recommendUsersForProducts and recommendProductsForUsers in matrix factorization model for PySpark
Author: Vladimir Vladimirov <vladimir.vladimirov@magnetic.com>
Closes#8700 from smartkiwi/SPARK-10535_.
Compute upper triangular values of the covariance matrix, then copy to lower triangular values.
Author: Nick Pritchard <nicholas.pritchard@falkonry.com>
Closes#8940 from pnpritchard/SPARK-10875.
GBT compare ValidateError with tolerance switching between relative and absolute ones, where the former one is relative to the current loss on the training set.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8549 from yanboliang/spark-7770.
LinearRegression training summary: The transformed dataset should hold all columns, not just selected ones like prediction and label. There is no real need to remove some, and the user may find them useful.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8564 from holdenk/SPARK-9718-LinearRegressionTrainingSummary-all-columns.
Reimplement `DecisionTree.findSplitsBins` via `RDD` to parallelize bin calculation.
With large feature spaces the current implementation is very slow. This change limits the features that are distributed (or collected) to just the continuous features, and performs the split calculations in parallel. It completes on a real multi terabyte dataset in less than a minute instead of multiple hours.
Author: Nathan Howell <nhowell@godaddy.com>
Closes#8246 from NathanHowell/SPARK-10064.
Refactoring `Instance` case class out from LOR and LIR, and also cleaning up some code.
Author: DB Tsai <dbt@netflix.com>
Closes#8853 from dbtsai/refactoring.
Provide initialModel param for pyspark.mllib.clustering.KMeans
Author: Evan Chen <chene@us.ibm.com>
Closes#8967 from evanyc15/SPARK-10779-pyspark-mllib.
It is currently impossible to clear Param values once set. It would be helpful to be able to.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8619 from holdenk/SPARK-9841-params-clear-needs-to-be-public.
JIRA issue [here](https://issues.apache.org/jira/browse/SPARK-5890).
I borrow the code of `findSplits` from `RandomForest`. I don't think it's good to call it from `RandomForest` directly.
Author: Xusen Yin <yinxusen@gmail.com>
Closes#5779 from yinxusen/SPARK-5890.
For some implicit dataset, ratings may not exist in the training data. In this case, we can assume all observed pairs to be positive and treat their ratings as 1. This should happen when users set ```ratingCol``` to an empty string.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8937 from yanboliang/spark-10736.
I implemented toString for AssociationRules.Rule, format like `[x, y] => {z}: 1.0`
Author: y-shimizu <y.shimizu0429@gmail.com>
Closes#8904 from y-shimizu/master.
This integrates the Interaction feature transformer with SparkR R formula support (i.e. support `:`).
To generate reasonable ML attribute names for feature interactions, it was necessary to add the ability to read attribute the original attribute names back from `StructField`, and also to specify custom group prefixes in `VectorAssembler`. This also has the side-benefit of cleaning up the double-underscores in the attributes generated for non-interaction terms.
mengxr
Author: Eric Liang <ekl@databricks.com>
Closes#8830 from ericl/interaction-2.
As introduced in https://issues.apache.org/jira/browse/SPARK-10630 we now have an easier way to create dataframes from local Java lists. Lets update the tests to use those.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8886 from holdenk/SPARK-10763-update-java-mllib-ml-tests-to-use-simplified-dataframe-construction.
Currently use can set ```checkpointInterval``` to specify how often should the cache be check-pointed. But we also need the function that users can disable it. This PR supports that users can disable checkpoint if user setting ```checkpointInterval = -1```.
We also add documents for GBT ```cacheNodeIds``` to make users can understand more clearly about checkpoint.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8820 from yanboliang/spark-10699.
By default ```quantilesCol``` should be empty. If ```quantileProbabilities``` is set, we should append quantiles as a new column (of type Vector).
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8836 from yanboliang/spark-10686.
All prediction models should store `numFeatures` indicating the number of features the model was trained on. Default value of -1 added for backwards compatibility.
Author: sethah <seth.hendrickson16@gmail.com>
Closes#8675 from sethah/SPARK-9715.
Currently when you set illegal value for params of array type (such as IntArrayParam, DoubleArrayParam, StringArrayParam), it will throw IllegalArgumentException but with incomprehensible error information.
Take ```VectorSlicer.setNames``` as an example:
```scala
val vectorSlicer = new VectorSlicer().setInputCol("features").setOutputCol("result")
// The value of setNames must be contain distinct elements, so the next line will throw exception.
vectorSlicer.setIndices(Array.empty).setNames(Array("f1", "f4", "f1"))
```
It will throw IllegalArgumentException as:
```
vectorSlicer_b3b4d1a10f43 parameter names given invalid value [Ljava.lang.String;798256c5.
java.lang.IllegalArgumentException: vectorSlicer_b3b4d1a10f43 parameter names given invalid value [Ljava.lang.String;798256c5.
```
We should distinguish the value of array type from primitive type at Param.validate(value: T), and we will get better error information.
```
vectorSlicer_3b744ea277b2 parameter names given invalid value [f1,f4,f1].
java.lang.IllegalArgumentException: vectorSlicer_3b744ea277b2 parameter names given invalid value [f1,f4,f1].
```
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8863 from yanboliang/spark-10750.
NodeIdCache: prevNodeIdsForInstances.unpersist() needs to be called at end of training.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8541 from holdenk/SPARK-9962-decission-tree-training-prevNodeIdsForiNstances-unpersist-at-end-of-training.
In many modeling application, data points are not necessarily sampled with equal probabilities. Linear regression should support weighting which account the over or under sampling.
work in progress.
Author: Meihua Wu <meihuawu@umich.edu>
Closes#8631 from rotationsymmetry/SPARK-9642.
SPARK-3136 added a large number of functions for creating Java RandomRDDs, but for people that want to use custom RandomDataGenerators we should make a Java friendly method.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8782 from holdenk/SPARK-10626-create-java-friendly-method-for-randomRDD.
There are duplicate set of initialization flag in `WeightedLeastSquares#add`.
`initialized` is already set in `init(Int)`.
Author: lewuathe <lewuathe@me.com>
Closes#8837 from Lewuathe/duplicate-initialization-flag.
Note methods that fail for cols > 65535; note that SVD does not require n >= m
CC mengxr
Author: Sean Owen <sowen@cloudera.com>
Closes#8839 from srowen/SPARK-5905.
This makes equality test failures much more readable.
mengxr
Author: Eric Liang <ekl@databricks.com>
Author: Eric Liang <ekhliang@gmail.com>
Closes#8826 from ericl/attrgroupstr.
```GBTParams``` has ```stepSize``` as learning rate currently.
ML has shared param class ```HasStepSize```, ```GBTParams``` can extend from it rather than duplicated implementation.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8552 from yanboliang/spark-10394.
Should be the same as SPARK-7808 but use Java for the code example.
It would be great to add package doc for `spark.ml.feature`.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8740 from holdenk/SPARK-10077-JAVA-PACKAGE-DOC-FOR-SPARK.ML.FEATURE.
In fraud detection dataset, almost all the samples are negative while only couple of them are positive. This type of high imbalanced data will bias the models toward negative resulting poor performance. In python-scikit, they provide a correction allowing users to Over-/undersample the samples of each class according to the given weights. In auto mode, selects weights inversely proportional to class frequencies in the training set. This can be done in a more efficient way by multiplying the weights into loss and gradient instead of doing actual over/undersampling in the training dataset which is very expensive.
http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html
On the other hand, some of the training data maybe more important like the training samples from tenure users while the training samples from new users maybe less important. We should be able to provide another "weight: Double" information in the LabeledPoint to weight them differently in the learning algorithm.
Author: DB Tsai <dbt@netflix.com>
Author: DB Tsai <dbt@dbs-mac-pro.corp.netflix.com>
Closes#7884 from dbtsai/SPARK-7685.
This change does two things:
- tag a few tests and adds the mechanism in the build to be able to disable those tags,
both in maven and sbt, for both junit and scalatest suites.
- add some logic to run-tests.py to disable some tags depending on what files have
changed; that's used to disable expensive tests when a module hasn't explicitly
been changed, to speed up testing for changes that don't directly affect those
modules.
Author: Marcelo Vanzin <vanzin@cloudera.com>
Closes#8437 from vanzin/test-tags.
jira: https://issues.apache.org/jira/browse/SPARK-10491
We implemented dspr with sparse vector support in `RowMatrix`. This method is also used in WeightedLeastSquares and other places. It would be useful to move it to `linalg.BLAS`.
Let me know if new UT needed.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#8663 from hhbyyh/movedspr.
Fixes bug where IndexToString output schema was DoubleType. Correct me if I'm wrong, but it doesn't seem like the output needs to have any "ML Attribute" metadata.
Author: Nick Pritchard <nicholas.pritchard@falkonry.com>
Closes#8751 from pnpritchard/SPARK-10573.
[SPARK-3382](https://issues.apache.org/jira/browse/SPARK-3382) added a ```convergenceTol``` parameter for GradientDescent-based methods in Scala. We need that parameter in Python; otherwise, Python users will not be able to adjust that behavior (or even reproduce behavior from previous releases since the default changed).
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8457 from yanboliang/spark-10194.
A few Identifiable types did override their toString method but without using the parent implementation. As a consequence, the uid was not present anymore in the toString result. It is the default behaviour.
This patch is a quick fix. The question of enforcement is still up.
No tests have been written to verify the toString method behaviour. That would be long to do because all types should be tested and not only those which have a regression now.
It is possible to enforce the condition using the compiler by making the toString method final but that would introduce unwanted potential API breaking changes (see jira).
Author: Bertrand Dechoux <BertrandDechoux@users.noreply.github.com>
Closes#8062 from BertrandDechoux/SPARK-9720.
Changes:
* Make Scala doc for StringIndexerInverse clearer. Also remove Scala doc from transformSchema, so that the doc is inherited.
* MetadataUtils.scala: “ Helper utilities for tree-based algorithms” —> not just trees anymore
CC: holdenk mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8679 from jkbradley/doc-fixes-1.5.
We should document options in public API doc. Otherwise, it is hard to find out the options without looking at the code. I tried to make `DefaultSource` private and put the documentation to package doc. However, since then there exists no public class under `source.libsvm`, the Java package doc doesn't show up in the generated html file (http://bugs.java.com/bugdatabase/view_bug.do?bug_id=4492654). So I put the doc to `DefaultSource` instead. There are several minor updates in this PR:
1. Do `vectorType == "sparse"` only once.
2. Update `hashCode` and `equals`.
3. Remove inherited doc.
4. Delete temp dir in `afterAll`.
Lewuathe
Author: Xiangrui Meng <meng@databricks.com>
Closes#8699 from mengxr/SPARK-10537.
"checkpointInterval" is member of DecisionTreeParams in Scala API which is inconsistency with Python API, we should unified them.
```
member of DecisionTreeParams <-> Scala API
shared param for all ML Transformer/Estimator <-> Python API
```
Proposal:
"checkpointInterval" is also used by ALS, so we make it shared params at Scala.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8528 from yanboliang/spark-10023.
It is convenient to implement data source API for LIBSVM format to have a better integration with DataFrames and ML pipeline API.
Two option is implemented.
* `numFeatures`: Specify the dimension of features vector
* `featuresType`: Specify the type of output vector. `sparse` is default.
Author: lewuathe <lewuathe@me.com>
Closes#8537 from Lewuathe/SPARK-10117 and squashes the following commits:
986999d [lewuathe] Change unit test phrase
11d513f [lewuathe] Fix some reviews
21600a4 [lewuathe] Merge branch 'master' into SPARK-10117
9ce63c7 [lewuathe] Rewrite service loader file
1fdd2df [lewuathe] Merge branch 'SPARK-10117' of github.com:Lewuathe/spark into SPARK-10117
ba3657c [lewuathe] Merge branch 'master' into SPARK-10117
0ea1c1c [lewuathe] LibSVMRelation is registered into META-INF
4f40891 [lewuathe] Improve test suites
5ab62ab [lewuathe] Merge branch 'master' into SPARK-10117
8660d0e [lewuathe] Fix Java unit test
b56a948 [lewuathe] Merge branch 'master' into SPARK-10117
2c12894 [lewuathe] Remove unnecessary tag
7d693c2 [lewuathe] Resolv conflict
62010af [lewuathe] Merge branch 'master' into SPARK-10117
a97ee97 [lewuathe] Fix some points
aef9564 [lewuathe] Fix
70ee4dd [lewuathe] Add Java test
3fd8dce [lewuathe] [SPARK-10117] Implement SQL data source API for reading LIBSVM data
40d3027 [lewuathe] Add Java test
7056d4a [lewuathe] Merge branch 'master' into SPARK-10117
99accaa [lewuathe] [SPARK-10117] Implement SQL data source API for reading LIBSVM data
The bulk of the changes are on `transient` annotation on class parameter. Often the compiler doesn't generate a field for this parameters, so the the transient annotation would be unnecessary.
But if the class parameter are used in methods, then fields are created. So it is safer to keep the annotations.
The remainder are some potential bugs, and deprecated syntax.
Author: Luc Bourlier <luc.bourlier@typesafe.com>
Closes#8433 from skyluc/issue/sbt-2.11.
Add WeibullGenerator for RandomDataGenerator.
#8611 need use WeibullGenerator to generate random data based on Weibull distribution.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8622 from yanboliang/spark-10464.
The goal of this PR is to have a weighted least squares implementation that takes the normal equation approach, and hence to be able to provide R-like summary statistics and support IRLS (used by GLMs). The tests match R's lm and glmnet.
There are couple TODOs that can be addressed in future PRs:
* consolidate summary statistics aggregators
* move `dspr` to `BLAS`
* etc
It would be nice to have this merged first because it blocks couple other features.
dbtsai
Author: Xiangrui Meng <meng@databricks.com>
Closes#8588 from mengxr/SPARK-9834.
Loader.checkSchema was called to verify the schema after dataframe.select(...).
Schema verification should be done before dataframe.select(...)
Author: Vinod K C <vinod.kc@huawei.com>
Closes#8636 from vinodkc/fix_GaussianMixtureModel_load_verification.
Copied model must have the same parent, but ml.IsotonicRegressionModel.copy did not set parent.
Here fix it and add test case.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8637 from yanboliang/spark-10470.
This PR fix two model ```copy()``` related issues:
[SPARK-10480](https://issues.apache.org/jira/browse/SPARK-10480)
```ML.LinearRegressionModel.copy()``` ignored argument ```extra```, it will not take effect when users setting this parameter.
[SPARK-10479](https://issues.apache.org/jira/browse/SPARK-10479)
```ML.LogisticRegressionModel.copy()``` should copy model summary if available.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8641 from yanboliang/linear-regression-copy.
From Jira: We should use assertTrue, etc. instead to make sure the asserts are not ignored in tests.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8607 from holdenk/SPARK-10013-remove-java-assert-from-java-unit-tests.
We should make sure the scaladoc for params includes their default values through the models in ml/
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8591 from holdenk/SPARK-10402-add-scaladoc-for-default-values-of-params-in-ml.
Params.getOrDefault should throw a more meaningful exception than what you get from a bad key lookup.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8567 from holdenk/SPARK-9723-params-getordefault-should-throw-more-useful-error.
Currently OneVsRest use UDF to generate new binary label during training.
Considering that [SPARK-7321](https://issues.apache.org/jira/browse/SPARK-7321) has been merged, we can use ```when ... otherwise``` which will be more efficiency.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8519 from yanboliang/spark-10349.
This could help reduce hash collisions, e.g., in `RDD[Vector].repartition`. jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#8182 from mengxr/SPARK-9954.
* do not cache first cost RDD
* change following cost RDD cache level to MEMORY_AND_DISK
* remove Vector wrapper to save a object per instance
Further improvements will be addressed in SPARK-10329
cc: yu-iskw HuJiayin
Author: Xiangrui Meng <meng@databricks.com>
Closes#8526 from mengxr/SPARK-10354.
* Adds user guide for ml.feature.StopWordsRemovers, ran code examples on my machine
* Cleans up scaladocs for public methods
* Adds test for Java compatibility
* Follow up Python user guide code example is tracked by SPARK-10249
Author: Feynman Liang <fliang@databricks.com>
Closes#8436 from feynmanliang/SPARK-10230.
`GeneralizedLinearModel` creates a cached RDD when building a model. It's inconvenient, since these RDDs flood the memory when building several models in a row, so useful data might get evicted from the cache.
The proposed solution is to always cache the dataset & remove the warning. There's a caveat though: input dataset gets evaluated twice, in line 270 when fitting `StandardScaler` for the first time, and when running optimizer for the second time. So, it might worth to return removed warning.
Another possible solution is to disable caching entirely & return removed warning. I don't really know what approach is better.
Author: Vyacheslav Baranov <slavik.baranov@gmail.com>
Closes#8395 from SlavikBaranov/SPARK-10182.
* Replaces instances of `Lists.newArrayList` with `Arrays.asList`
* Replaces `commons.lang.StringUtils` over `com.google.collections.Strings`
* Replaces `List` interface over `ArrayList` implementations
This PR along with #8445#8446#8447 completely removes all `com.google.collections.Lists` dependencies within mllib's Java tests.
Author: Feynman Liang <fliang@databricks.com>
Closes#8451 from feynmanliang/SPARK-10257.
Fix for [JavaConverters.asJavaListConverter](http://www.scala-lang.org/api/2.10.5/index.html#scala.collection.JavaConverters$) being removed in 2.11.7 and hence the build fails with the 2.11 profile enabled. Tested with the default 2.10 and 2.11 profiles. BUILD SUCCESS in both cases.
Build for 2.10:
./build/mvn -Pyarn -Phadoop-2.6 -Dhadoop.version=2.7.1 -DskipTests clean install
and 2.11:
./dev/change-scala-version.sh 2.11
./build/mvn -Pyarn -Phadoop-2.6 -Dhadoop.version=2.7.1 -Dscala-2.11 -DskipTests clean install
Author: Jacek Laskowski <jacek@japila.pl>
Closes#8479 from jaceklaskowski/SPARK-9613-hotfix.
* Replaces `com.google.common` dependencies with `java.util.Arrays`
* Small clean up in `JavaNormalizerSuite`
Author: Feynman Liang <fliang@databricks.com>
Closes#8445 from feynmanliang/SPARK-10254.
I only found `ml.NaiveBayes` missing `Experimental` annotation. This PR doesn't cover Python APIs.
cc jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#8452 from mengxr/SPARK-9665.
Same as #8421 but for `mllib.feature`.
cc dbtsai
Author: Xiangrui Meng <meng@databricks.com>
Closes#8449 from mengxr/SPARK-10236.feature and squashes the following commits:
0e8d658 [Xiangrui Meng] remove unnecessary comment
ad70b03 [Xiangrui Meng] update since versions in mllib.feature
Same as #8421 but for `mllib.regression`.
cc freeman-lab dbtsai
Author: Xiangrui Meng <meng@databricks.com>
Closes#8426 from mengxr/SPARK-10235 and squashes the following commits:
6cd28e4 [Xiangrui Meng] update since versions in mllib.regression
The same as #8241 but for `mllib.stat` and `mllib.random`.
cc feynmanliang
Author: Xiangrui Meng <meng@databricks.com>
Closes#8439 from mengxr/SPARK-10242.
Same as #8421 but for `mllib.linalg`.
cc dbtsai
Author: Xiangrui Meng <meng@databricks.com>
Closes#8440 from mengxr/SPARK-10238 and squashes the following commits:
b38437e [Xiangrui Meng] update since versions in mllib.linalg
* Adds two new sections to LDA's user guide; one for each optimizer/model
* Documents new features added to LDA (e.g. topXXXperXXX, asymmetric priors, hyperpam optimization)
* Cleans up a TODO and sets a default parameter in LDA code
jkbradley hhbyyh
Author: Feynman Liang <fliang@databricks.com>
Closes#8254 from feynmanliang/SPARK-9888.
Same as #8421 but for `mllib.pmml` and `mllib.util`.
cc dbtsai
Author: Xiangrui Meng <meng@databricks.com>
Closes#8430 from mengxr/SPARK-10239 and squashes the following commits:
a189acf [Xiangrui Meng] update since versions in mllib.pmml and mllib.util
Adds default convergence tolerance (0.001, set in `GradientDescent.convergenceTol`) to `setConvergenceTol`'s scaladoc
Author: Feynman Liang <fliang@databricks.com>
Closes#8424 from feynmanliang/SPARK-9797.
* Adds doc for alias of runMIniBatchSGD documenting default value for convergeTol
* Cleans up a note in code
Author: Feynman Liang <fliang@databricks.com>
Closes#8425 from feynmanliang/SPARK-9800.
Update `Since` annotation in `mllib.classification`:
1. add version to classes, objects, constructors, and public variables declared in constructors
2. correct some versions
3. remove `Since` on `toString`
MechCoder dbtsai
Author: Xiangrui Meng <meng@databricks.com>
Closes#8421 from mengxr/SPARK-10231 and squashes the following commits:
b2dce80 [Xiangrui Meng] update @Since annotation for mllib.classification
Replace `JavaConversions` implicits with `JavaConverters`
Most occurrences I've seen so far are necessary conversions; a few have been avoidable. None are in critical code as far as I see, yet.
Author: Sean Owen <sowen@cloudera.com>
Closes#8033 from srowen/SPARK-9613.
GaussianMixture now distributes matrix decompositions for certain problem sizes. Distributed computation actually fails, but this was not tested in unit tests.
This PR adds a unit test which checks this. It failed previously but works with this fix.
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8370 from jkbradley/gmm-fix.
Add user guide for `VectorSlicer`, with Java test suite and Python version VectorSlicer.
Note that Python version does not support selecting by names now.
Author: Xusen Yin <yinxusen@gmail.com>
Closes#8267 from yinxusen/SPARK-9893.
Removed categorical feature info validation since no longer needed
This is needed to make the ML user guide examples work (in another current PR).
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8367 from jkbradley/gbt-single-cat.
For each (document, term) pair, return top topic. Note that instances of (doc, term) pairs within a document (a.k.a. "tokens") are exchangeable, so we should provide an estimate per document-term, rather than per token.
CC: rotationsymmetry mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8329 from jkbradley/lda-topic-assignments.
This continues the work from #8256. I removed `since` tags from private/protected/local methods/variables (see 72fdeb6463). MechCoder
Closes#8256
Author: Xiangrui Meng <meng@databricks.com>
Author: Xiaoqing Wang <spark445@126.com>
Author: MechCoder <manojkumarsivaraj334@gmail.com>
Closes#8288 from mengxr/SPARK-8918.
Previously, users of evaluator (`CrossValidator` and `TrainValidationSplit`) would only maximize the metric in evaluator, leading to a hacky solution which negated metrics to be minimized and caused erroneous negative values to be reported to the user.
This PR adds a `isLargerBetter` attribute to the `Evaluator` base class, instructing users of `Evaluator` on whether the chosen metric should be maximized or minimized.
CC jkbradley
Author: Feynman Liang <fliang@databricks.com>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8290 from feynmanliang/SPARK-10097.
jira: https://issues.apache.org/jira/browse/SPARK-9028
Add an estimator for CountVectorizerModel. The estimator will extract a vocabulary from document collections according to the term frequency.
I changed the meaning of minCount as a filter across the corpus. This aligns with Word2Vec and the similar parameter in SKlearn.
Author: Yuhao Yang <hhbyyh@gmail.com>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#7388 from hhbyyh/cvEstimator.
Fix the issue that ```layers``` and ```weights``` should be public variables of ```MultilayerPerceptronClassificationModel```. Users can not get ```layers``` and ```weights``` from a ```MultilayerPerceptronClassificationModel``` currently.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8263 from yanboliang/mlp-public.
This PR adds a short description of `ml.feature` package with code example. The Java package doc will come in a separate PR. jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#8260 from mengxr/SPARK-7808.
Added since tags to mllib.regression
Author: Prayag Chandran <prayagchandran@gmail.com>
Closes#7518 from prayagchandran/sinceTags and squashes the following commits:
fa4dda2 [Prayag Chandran] Re-formatting
6c6d584 [Prayag Chandran] Corrected a few tags. Removed few unnecessary tags
1a0365f [Prayag Chandran] Reformating and adding a few more tags
89fdb66 [Prayag Chandran] SPARK-8916 [Documentation, MLlib] Add @since tags to mllib.regression
Also added unit test for integration between StringIndexerModel and IndexToString
CC: holdenk We realized we should have left in your unit test (to catch the issue with removing the inverse() method), so this adds it back. mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8211 from jkbradley/stridx-labels.
in MLlib sometimes we need to set metadata for the new column, thus we will alias the new column with metadata before call `withColumn` and in `withColumn` we alias this clolumn again. Here I overloaded `withColumn` to allow user set metadata, just like what we did for `Column.as`.
Author: Wenchen Fan <cloud0fan@outlook.com>
Closes#8159 from cloud-fan/withColumn.
It would be helpful to allow users to pass a pre-computed index to create an indexer, rather than always going through StringIndexer to create the model.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#7267 from holdenk/SPARK-8744-StringIndexerModel-should-have-public-constructor.
This modifies DecisionTreeMetadata construction to treat 1-category features as continuous, so that trees do not fail with such features. It is important for the pipelines API, where VectorIndexer can automatically categorize certain features as categorical.
As stated in the JIRA, this is a temp fix which we can improve upon later by automatically filtering out those features. That will take longer, though, since it will require careful indexing.
Targeted for 1.5 and master
CC: manishamde mengxr yanboliang
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8187 from jkbradley/tree-1cat.
Some minor clean-ups after SPARK-9661. See my inline comments. MechCoder jkbradley
Author: Xiangrui Meng <meng@databricks.com>
Closes#8190 from mengxr/SPARK-9661-fix.
What `StringIndexerInverse` does is not strictly associated with `StringIndexer`, and the name is not clearly describing the transformation. Renaming to `IndexToString` might be better.
~~I also changed `invert` to `inverse` without arguments. `inputCol` and `outputCol` could be set after.~~
I also removed `invert`.
jkbradley holdenk
Author: Xiangrui Meng <meng@databricks.com>
Closes#8152 from mengxr/SPARK-9922.
I skimmed through the docs for various instance of Object and replaced them with Java compaible versions of the same.
1. Some methods in LDAModel.
2. runMiniBatchSGD
3. kolmogorovSmirnovTest
Author: MechCoder <manojkumarsivaraj334@gmail.com>
Closes#8126 from MechCoder/java_incop.
To follow the naming rule of ML, change `MultilayerPerceptronClassifierModel` to `MultilayerPerceptronClassificationModel` like `DecisionTreeClassificationModel`, `GBTClassificationModel` and so on.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#8164 from yanboliang/mlp-name.
Copied ML models must have the same parent of original ones
Author: lewuathe <lewuathe@me.com>
Author: Lewuathe <lewuathe@me.com>
Closes#7447 from Lewuathe/SPARK-9073.
This requires some discussion. I'm not sure whether `runs` is a useful parameter. It certainly complicates the implementation. We might want to optimize the k-means implementation with block matrix operations. In this case, having `runs` may not be worth the trade-off. Also it increases the communication cost in a single job, which might cause other issues.
This PR also renames `epsilon` to `tol` to have consistent naming among algorithms. The Python constructor is updated to include all parameters.
jkbradley yu-iskw
Author: Xiangrui Meng <meng@databricks.com>
Closes#8148 from mengxr/SPARK-9918 and squashes the following commits:
149b9e5 [Xiangrui Meng] fix constructor in Python and rename epsilon to tol
3cc15b3 [Xiangrui Meng] fix test and change initStep to initSteps in python
a0a0274 [Xiangrui Meng] remove runs from k-means in the pipeline API
The problem with defining setters in the base class is that it doesn't return the correct type in Java.
ericl
Author: Xiangrui Meng <meng@databricks.com>
Closes#8143 from mengxr/SPARK-9914 and squashes the following commits:
d36c887 [Xiangrui Meng] remove setters from model
a49021b [Xiangrui Meng] define setters explicitly for Java and use setParam group
There exists a chance that the prefixes keep growing to the maximum pattern length. Then the final local processing step becomes unnecessary. feynmanliang
Author: Xiangrui Meng <meng@databricks.com>
Closes#8136 from mengxr/SPARK-9903.
Made ProbabilisticClassifier, Identifiable, VectorUDT public. All are annotated as DeveloperApi.
CC: mengxr EronWright
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8004 from jkbradley/ml-api-public-items and squashes the following commits:
7ebefda [Joseph K. Bradley] update per code review
7ff0768 [Joseph K. Bradley] attepting to add mima fix
756d84c [Joseph K. Bradley] VectorUDT annotated as AlphaComponent
ae7767d [Joseph K. Bradley] added another warning
94fd553 [Joseph K. Bradley] Made ProbabilisticClassifier, Identifiable, VectorUDT public APIs
As per the TODO move weightCol to Shared Params.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#8144 from holdenk/SPARK-9909-move-weightCol-toSharedParams.
Reinstated LogisticRegression.threshold Param for binary compatibility. Param thresholds overrides threshold, if set.
CC: mengxr dbtsai feynmanliang
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8079 from jkbradley/logreg-reinstate-threshold.
From JIRA: Currently, Params.copyValues copies default parameter values to the paramMap of the target instance, rather than the defaultParamMap. It should copy to the defaultParamMap because explicitly setting a parameter can change the semantics.
This issue arose in SPARK-9789, where 2 params "threshold" and "thresholds" for LogisticRegression can have mutually exclusive values. If thresholds is set, then fit() will copy the default value of threshold as well, easily resulting in inconsistent settings for the 2 params.
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#8115 from jkbradley/copyvalues-fix.
Went thru the history of changes the file MLUtils.scala and picked up the version that the change went in.
Author: Sudhakar Thota <sudhakarthota@yahoo.com>
Author: Sudhakar Thota <sudhakarthota@sudhakars-mbp-2.usca.ibm.com>
Closes#7436 from sthota2014/SPARK-8925_thotas.
1. Add “asymmetricDocConcentration” and revert docConcentration changes. If the (internal) doc concentration vector is a single value, “getDocConcentration" returns it. If it is a constant vector, getDocConcentration returns the first item, and fails otherwise.
2. Give `LDAModel.gammaShape` a default value in `LDAModel` concrete class constructors.
jkbradley
Author: Feynman Liang <fliang@databricks.com>
Closes#8077 from feynmanliang/SPARK-9788 and squashes the following commits:
6b07bc8 [Feynman Liang] Code review changes
9d6a71e [Feynman Liang] Add asymmetricAlpha alias
bf4e685 [Feynman Liang] Asymmetric docConcentration
4cab972 [Feynman Liang] Default gammaShape
Adds unit test for `equals` on `mllib.linalg.Matrix` class and `equals` to both `SparseMatrix` and `DenseMatrix`. Supports equality testing between `SparseMatrix` and `DenseMatrix`.
mengxr
Author: Feynman Liang <fliang@databricks.com>
Closes#8042 from feynmanliang/SPARK-9750 and squashes the following commits:
bb70d5e [Feynman Liang] Breeze compare for dense matrices as well, in case other is sparse
ab6f3c8 [Feynman Liang] Sparse matrix compare for equals
22782df [Feynman Liang] Add equality based on matrix semantics, not representation
78f9426 [Feynman Liang] Add casts
43d28fa [Feynman Liang] Fix failing test
6416fa0 [Feynman Liang] Add failing sparse matrix equals tests
As a precursor to adding a public constructor add an option to handle unseen values by skipping rather than throwing an exception (default remains throwing an exception),
Author: Holden Karau <holden@pigscanfly.ca>
Closes#7266 from holdenk/SPARK-8764-string-indexer-should-take-option-to-handle-unseen-values and squashes the following commits:
38a4de9 [Holden Karau] fix long line
045bf22 [Holden Karau] Add a second b entry so b gets 0 for sure
81dd312 [Holden Karau] Update the docs for handleInvalid param to be more descriptive
7f37f6e [Holden Karau] remove extra space (scala style)
414e249 [Holden Karau] And switch to using handleInvalid instead of skipInvalid
1e53f9b [Holden Karau] update the param (codegen side)
7a22215 [Holden Karau] fix typo
100a39b [Holden Karau] Merge in master
aa5b093 [Holden Karau] Since we filter we should never go down this code path if getSkipInvalid is true
75ffa69 [Holden Karau] Remove extra newline
d69ef5e [Holden Karau] Add a test
b5734be [Holden Karau] Add support for unseen labels
afecd4e [Holden Karau] Add a param to skip invalid entries.
Implements the transforms which are defined by SQL statement.
Currently we only support SQL syntax like 'SELECT ... FROM __THIS__'
where '__THIS__' represents the underlying table of the input dataset.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#7465 from yanboliang/spark-8345 and squashes the following commits:
b403fcb [Yanbo Liang] address comments
0d4bb15 [Yanbo Liang] a better transformSchema() implementation
51eb9e7 [Yanbo Liang] Add an SQL node as a feature transformer
Adds method documentations back to `MultivariateOnlineSummarizer`, which were present in 1.4 but disappeared somewhere along the way to 1.5.
jkbradley
Author: Feynman Liang <fliang@databricks.com>
Closes#8045 from feynmanliang/SPARK-9755 and squashes the following commits:
af67fde [Feynman Liang] Add MultivariateOnlineSummarizer docs
Small documentation cleanups, including:
* Adds documentation for `pi` and `theta`
* setParam to `setModelType`
Author: Feynman Liang <fliang@databricks.com>
Closes#8047 from feynmanliang/SPARK-9719 and squashes the following commits:
b372438 [Feynman Liang] Clean up naive bayes doc
These should be made private until there is a public constructor for providing `rootNode: Node` to use these constructors.
jkbradley
Author: Feynman Liang <fliang@databricks.com>
Closes#8046 from feynmanliang/SPARK-9756 and squashes the following commits:
2cbdf08 [Feynman Liang] Make RFRegressionModel aux constructor private
a06f596 [Feynman Liang] Make constructors in ML decision trees private
A minor typo (centriod -> centroid). Readable variable names help every users.
Author: Bertrand Dechoux <BertrandDechoux@users.noreply.github.com>
Closes#8037 from BertrandDechoux/kmeans-typo and squashes the following commits:
47632fe [Bertrand Dechoux] centriod typo
Resubmit of [https://github.com/apache/spark/pull/6906] for adding single-vec predict to GMMs
CC: dkobylarz mengxr
To be merged with master and branch-1.5
Primary author: dkobylarz
Author: Dariusz Kobylarz <darek.kobylarz@gmail.com>
Closes#8039 from jkbradley/gmm-predict-vec and squashes the following commits:
bfbedc4 [Dariusz Kobylarz] [SPARK-8481] [MLlib] GaussianMixtureModel predict accepting single vector
This PR contains the following changes:
* add `featureIndex` to handle vector features (in order to chain isotonic regression easily with output from logistic regression
* make getter/setter names consistent with params
* remove inheritance from Regressor because it is tricky to handle both `DoubleType` and `VectorType`
* simplify test data generation
jkbradley zapletal-martin
Author: Xiangrui Meng <meng@databricks.com>
Closes#7952 from mengxr/SPARK-9493 and squashes the following commits:
8818ac3 [Xiangrui Meng] address comments
05e2216 [Xiangrui Meng] address comments
8d08090 [Xiangrui Meng] add featureIndex to handle vector features make getter/setter names consistent with params remove inheritance from Regressor
After https://github.com/apache/spark/pull/7263 it is pretty straightforward to Python wrappers.
Author: MechCoder <manojkumarsivaraj334@gmail.com>
Closes#7930 from MechCoder/spark-9533 and squashes the following commits:
1bea394 [MechCoder] make getVectors a lazy val
5522756 [MechCoder] [SPARK-9533] [PySpark] [ML] Add missing methods in Word2Vec ML
I have added support for stats in LogisticRegression. The API is similar to that of LinearRegression with LogisticRegressionTrainingSummary and LogisticRegressionSummary
I have some queries and asked them inline.
Author: MechCoder <manojkumarsivaraj334@gmail.com>
Closes#7538 from MechCoder/log_reg_stats and squashes the following commits:
2e9f7c7 [MechCoder] Change defs into lazy vals
d775371 [MechCoder] Clean up class inheritance
9586125 [MechCoder] Add abstraction to handle Multiclass Metrics
40ad8ef [MechCoder] minor
640376a [MechCoder] remove unnecessary dataframe stuff and add docs
80d9954 [MechCoder] Added tests
fbed861 [MechCoder] DataFrame support for metrics
70a0fc4 [MechCoder] [SPARK-9112] [ML] Implement Stats for LogisticRegression
Add VectorSlicer transformer to spark.ml, with features specified as either indices or names. Transfers feature attributes for selected features.
Updated version of [https://github.com/apache/spark/pull/5731]
CC: yinxusen This updates your PR. You'll still be the primary author of this PR.
CC: mengxr
Author: Xusen Yin <yinxusen@gmail.com>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#7972 from jkbradley/yinxusen-SPARK-5895 and squashes the following commits:
b16e86e [Joseph K. Bradley] fixed scala style
71c65d2 [Joseph K. Bradley] fix import order
86e9739 [Joseph K. Bradley] cleanups per code review
9d8d6f1 [Joseph K. Bradley] style fix
83bc2e9 [Joseph K. Bradley] Updated VectorSlicer
98c6939 [Xusen Yin] fix style error
ecbf2d3 [Xusen Yin] change interfaces and params
f6be302 [Xusen Yin] Merge branch 'master' into SPARK-5895
e4781f2 [Xusen Yin] fix commit error
fd154d7 [Xusen Yin] add test suite of vector slicer
17171f8 [Xusen Yin] fix slicer
9ab9747 [Xusen Yin] add vector slicer
aa5a0bf [Xusen Yin] add vector slicer
mengxr
Author: Feynman Liang <fliang@databricks.com>
Closes#7974 from feynmanliang/SPARK-9657 and squashes the following commits:
7ca533f [Feynman Liang] Fix return type of getMaxPatternLength
mengxr This adds the `BlockMatrix` to PySpark. I have the conversions to `IndexedRowMatrix` and `CoordinateMatrix` ready as well, so once PR #7554 is completed (which relies on PR #7746), this PR can be finished.
Author: Mike Dusenberry <mwdusenb@us.ibm.com>
Closes#7761 from dusenberrymw/SPARK-6486_Add_BlockMatrix_to_PySpark and squashes the following commits:
27195c2 [Mike Dusenberry] Adding one more check to _convert_to_matrix_block_tuple, and a few minor documentation changes.
ae50883 [Mike Dusenberry] Minor update: BlockMatrix should inherit from DistributedMatrix.
b8acc1c [Mike Dusenberry] Moving BlockMatrix to pyspark.mllib.linalg.distributed, updating the logic to match that of the other distributed matrices, adding conversions, and adding documentation.
c014002 [Mike Dusenberry] Using properties for better documentation.
3bda6ab [Mike Dusenberry] Adding documentation.
8fb3095 [Mike Dusenberry] Small cleanup.
e17af2e [Mike Dusenberry] Adding BlockMatrix to PySpark.
This is a major refactoring of the PrefixSpan implementation. It contains the following changes:
1. Expand prefix with one item at a time. The existing implementation generates all subsets for each itemset, which might have scalability issue when the itemset is large.
2. Use a new internal format. `<(12)(31)>` is represented by `[0, 1, 2, 0, 1, 3, 0]` internally. We use `0` because negative numbers are used to indicates partial prefix items, e.g., `_2` is represented by `-2`.
3. Remember the start indices of all partial projections in the projected postfix to help next projection.
4. Reuse the original sequence array for projected postfixes.
5. Use `Prefix` IDs in aggregation rather than its content.
6. Use `ArrayBuilder` for building primitive arrays.
7. Expose `maxLocalProjDBSize`.
8. Tests are not changed except using `0` instead of `-1` as the delimiter.
`Postfix`'s API doc should be a good place to start.
Closes#7594
feynmanliang zhangjiajin
Author: Xiangrui Meng <meng@databricks.com>
Closes#7937 from mengxr/SPARK-9540 and squashes the following commits:
2d0ec31 [Xiangrui Meng] address more comments
48f450c [Xiangrui Meng] address comments from Feynman; fixed a bug in project and added a test
65f90e8 [Xiangrui Meng] naming and documentation
8afc86a [Xiangrui Meng] refactor impl
All compressed sensing applications, and some of the regression use-cases will have better result by turning the feature scaling off. However, if we implement this naively by training the dataset without doing any standardization, the rate of convergency will not be good. This can be implemented by still standardizing the training dataset but we penalize each component differently to get effectively the same objective function but a better numerical problem. As a result, for those columns with high variances, they will be penalized less, and vice versa. Without this, since all the features are standardized, so they will be penalized the same.
In R, there is an option for this.
standardize
Logical flag for x variable standardization, prior to fitting the model sequence. The coefficients are always returned on the original scale. Default is standardize=TRUE. If variables are in the same units already, you might not wish to standardize. See details below for y standardization with family="gaussian".
Note that the primary author for this PR is holdenk
Author: Holden Karau <holden@pigscanfly.ca>
Author: DB Tsai <dbt@netflix.com>
Closes#7875 from dbtsai/SPARK-8522 and squashes the following commits:
e856036 [DB Tsai] scala doc
596e96c [DB Tsai] minor
bbff347 [DB Tsai] naming
baa0805 [DB Tsai] touch up
d6234ba [DB Tsai] Merge branch 'master' into SPARK-8522-Disable-Linear_featureScaling-Spark-8601-in-Linear_regression
6b1dc09 [Holden Karau] Merge branch 'master' into SPARK-8522-Disable-Linear_featureScaling-Spark-8601-in-Linear_regression
332f140 [Holden Karau] Merge in master
eebe10a [Holden Karau] Use same comparision operator throughout the test
3f92935 [Holden Karau] merge
b83a41e [Holden Karau] Expand the tests and make them similar to the other PR also providing an option to disable standardization (but for LoR).
0c334a2 [Holden Karau] Remove extra line
99ce053 [Holden Karau] merge in master
e54a8a9 [Holden Karau] Fix long line
e47c574 [Holden Karau] Add support for L2 without standardization.
55d3a66 [Holden Karau] Add standardization param for linear regression
00a1dc5 [Holden Karau] Add the param to the linearregression impl
Update BinaryClassificationEvaluator to use setRawPredictionCol, rather than setScoreCol. Deprecated setScoreCol.
I don't think setScoreCol was actually used anywhere (based on search).
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#7921 from jkbradley/binary-eval-rawpred and squashes the following commits:
e5d7dfa [Joseph K. Bradley] Update BinaryClassificationEvaluator to use setRawPredictionCol
This PR adds the RowMatrix, IndexedRowMatrix, and CoordinateMatrix distributed matrices to PySpark. Each distributed matrix class acts as a wrapper around the Scala/Java counterpart by maintaining a reference to the Java object. New distributed matrices can be created using factory methods added to DistributedMatrices, which creates the Java distributed matrix and then wraps it with the corresponding PySpark class. This design allows for simple conversion between the various distributed matrices, and lets us re-use the Scala code. Serialization between Python and Java is implemented using DataFrames as needed for IndexedRowMatrix and CoordinateMatrix for simplicity. Associated documentation and unit-tests have also been added. To facilitate code review, this PR implements access to the rows/entries as RDDs, the number of rows & columns, and conversions between the various distributed matrices (not including BlockMatrix), and does not implement the other linear algebra functions of the matrices, although this will be very simple to add now.
Author: Mike Dusenberry <mwdusenb@us.ibm.com>
Closes#7554 from dusenberrymw/SPARK-6485_Add_CoordinateMatrix_RowMatrix_IndexedMatrix_to_PySpark and squashes the following commits:
bb039cb [Mike Dusenberry] Minor documentation update.
b887c18 [Mike Dusenberry] Updating the matrix conversion logic again to make it even cleaner. Now, we allow the 'rows' parameter in the constructors to be either an RDD or the Java matrix object. If 'rows' is an RDD, we create a Java matrix object, wrap it, and then store that. If 'rows' is a Java matrix object of the correct type, we just wrap and store that directly. This is only for internal usage, and publicly, we still require 'rows' to be an RDD. We no longer store the 'rows' RDD, and instead just compute it from the Java object when needed. The point of this is that when we do matrix conversions, we do the conversion on the Scala/Java side, which returns a Java object, so we should use that directly, but exposing 'java_matrix' parameter in the public API is not ideal. This non-public feature of allowing 'rows' to be a Java matrix object is documented in the '__init__' constructor docstrings, which are not part of the generated public API, and doctests are also included.
7f0dcb6 [Mike Dusenberry] Updating module docstring.
cfc1be5 [Mike Dusenberry] Use 'new SQLContext(matrix.rows.sparkContext)' rather than 'SQLContext.getOrCreate', as the later doesn't guarantee that the SparkContext will be the same as for the matrix.rows data.
687e345 [Mike Dusenberry] Improving conversion performance. This adds an optional 'java_matrix' parameter to the constructors, and pulls the conversion logic out into a '_create_from_java' function. Now, if the constructors are given a valid Java distributed matrix object as 'java_matrix', they will store those internally, rather than create a new one on the Scala/Java side.
3e50b6e [Mike Dusenberry] Moving the distributed matrices to pyspark.mllib.linalg.distributed.
308f197 [Mike Dusenberry] Using properties for better documentation.
1633f86 [Mike Dusenberry] Minor documentation cleanup.
f0c13a7 [Mike Dusenberry] CoordinateMatrix should inherit from DistributedMatrix.
ffdd724 [Mike Dusenberry] Updating doctests to make documentation cleaner.
3fd4016 [Mike Dusenberry] Updating docstrings.
27cd5f6 [Mike Dusenberry] Simplifying input conversions in the constructors for each distributed matrix.
a409cf5 [Mike Dusenberry] Updating doctests to be less verbose by using lists instead of DenseVectors explicitly.
d19b0ba [Mike Dusenberry] Updating code and documentation to note that a vector-like object (numpy array, list, etc.) can be used in place of explicit Vector object, and adding conversions when necessary to RowMatrix construction.
4bd756d [Mike Dusenberry] Adding param documentation to IndexedRow and MatrixEntry.
c6bded5 [Mike Dusenberry] Move conversion logic from tuples to IndexedRow or MatrixEntry types from within the IndexedRowMatrix and CoordinateMatrix constructors to separate _convert_to_indexed_row and _convert_to_matrix_entry functions.
329638b [Mike Dusenberry] Moving the Experimental tag to the top of each docstring.
0be6826 [Mike Dusenberry] Simplifying doctests by removing duplicated rows/entries RDDs within the various tests.
c0900df [Mike Dusenberry] Adding the colons that were accidentally not inserted.
4ad6819 [Mike Dusenberry] Documenting the and parameters.
3b854b9 [Mike Dusenberry] Minor updates to documentation.
10046e8 [Mike Dusenberry] Updating documentation to use class constructors instead of the removed DistributedMatrices factory methods.
119018d [Mike Dusenberry] Adding static methods to each of the distributed matrix classes to consolidate conversion logic.
4d7af86 [Mike Dusenberry] Adding type checks to the constructors. Although it is slightly verbose, it is better for the user to have a good error message than a cryptic stacktrace.
93b6a3d [Mike Dusenberry] Pulling the DistributedMatrices Python class out of this pull request.
f6f3c68 [Mike Dusenberry] Pulling the DistributedMatrices Scala class out of this pull request.
6a3ecb7 [Mike Dusenberry] Updating pattern matching.
08f287b [Mike Dusenberry] Slight reformatting of the documentation.
a245dc0 [Mike Dusenberry] Updating Python doctests for compatability between Python 2 & 3. Since Python 3 removed the idea of a separate 'long' type, all values that would have been outputted as a 'long' (ex: '4L') will now be treated as an 'int' and outputed as one (ex: '4'). The doctests now explicitly convert to ints so that both Python 2 and 3 will have the same output. This is fine since the values are all small, and thus can be easily represented as ints.
4d3a37e [Mike Dusenberry] Reformatting a few long Python doctest lines.
7e3ca16 [Mike Dusenberry] Fixing long lines.
f721ead [Mike Dusenberry] Updating documentation for each of the distributed matrices.
ab0e8b6 [Mike Dusenberry] Updating unit test to be more useful.
dda2f89 [Mike Dusenberry] Added wrappers for the conversions between the various distributed matrices. Added logic to be able to access the rows/entries of the distributed matrices, which requires serialization through DataFrames for IndexedRowMatrix and CoordinateMatrix types. Added unit tests.
0cd7166 [Mike Dusenberry] Implemented the CoordinateMatrix API in PySpark, following the idea of the IndexedRowMatrix API, including using DataFrames for serialization.
3c369cb [Mike Dusenberry] Updating the architecture a bit to make conversions between the various distributed matrix types easier. The different distributed matrix classes are now only wrappers around the Java objects, and take the Java object as an argument during construction. This way, we can call for example on an , which returns a reference to a Java RowMatrix object, and then construct a PySpark RowMatrix object wrapped around the Java object. This is analogous to the behavior of PySpark RDDs and DataFrames. We now delegate creation of the various distributed matrices from scratch in PySpark to the factory methods on .
4bdd09b [Mike Dusenberry] Implemented the IndexedRowMatrix API in PySpark, following the idea of the RowMatrix API. Note that for the IndexedRowMatrix, we use DataFrames to serialize the data between Python and Scala/Java, so we accept PySpark RDDs, then convert to a DataFrame, then convert back to RDDs on the Scala/Java side before constructing the IndexedRowMatrix.
23bf1ec [Mike Dusenberry] Updating documentation to add PySpark RowMatrix. Inserting newline above doctest so that it renders properly in API docs.
b194623 [Mike Dusenberry] Updating design to have a PySpark RowMatrix simply create and keep a reference to a wrapper over a Java RowMatrix. Updating DistributedMatrices factory methods to accept numRows and numCols with default values. Updating PySpark DistributedMatrices factory method to simply create a PySpark RowMatrix. Adding additional doctests for numRows and numCols parameters.
bc2d220 [Mike Dusenberry] Adding unit tests for RowMatrix methods.
d7e316f [Mike Dusenberry] Implemented the RowMatrix API in PySpark by doing the following: Added a DistributedMatrices class to contain factory methods for creating the various distributed matrices. Added a factory method for creating a RowMatrix from an RDD of Vectors. Added a createRowMatrix function to the PythonMLlibAPI to interface with the factory method. Added DistributedMatrix, DistributedMatrices, and RowMatrix classes to the pyspark.mllib.linalg api.
Small cleanups to recent LDA additions and docs.
CC: feynmanliang
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#7916 from jkbradley/lda-cleanups and squashes the following commits:
f7021d9 [Joseph K. Bradley] broadcasting large matrices for LDA in local model and online learning
97947aa [Joseph K. Bradley] a few more cleanups
5b03f88 [Joseph K. Bradley] reverted split of lda log likelihood
c566915 [Joseph K. Bradley] small edit to make review easier
63f6c7d [Joseph K. Bradley] clarified log likelihood for lda models
This PR replaces the old "threshold" with a generalized "thresholds" Param. We keep getThreshold,setThreshold for backwards compatibility for binary classification.
Note that the primary author of this PR is holdenk
Author: Holden Karau <holden@pigscanfly.ca>
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#7909 from jkbradley/holdenk-SPARK-8069-add-cutoff-aka-threshold-to-random-forest and squashes the following commits:
3952977 [Joseph K. Bradley] fixed pyspark doc test
85febc8 [Joseph K. Bradley] made python unit tests a little more robust
7eb1d86 [Joseph K. Bradley] small cleanups
6cc2ed8 [Joseph K. Bradley] Fixed remaining merge issues.
0255e44 [Joseph K. Bradley] Many cleanups for thresholds, some more tests
7565a60 [Holden Karau] fix pep8 style checks, add a getThreshold method similar to our LogisticRegression.scala one for API compat
be87f26 [Holden Karau] Convert threshold to thresholds in the python code, add specialized support for Array[Double] to shared parems codegen, etc.
6747dad [Holden Karau] Override raw2prediction for ProbabilisticClassifier, fix some tests
25df168 [Holden Karau] Fix handling of thresholds in LogisticRegression
c02d6c0 [Holden Karau] No default for thresholds
5e43628 [Holden Karau] CR feedback and fixed the renamed test
f3fbbd1 [Holden Karau] revert the changes to random forest :(
51f581c [Holden Karau] Add explicit types to public methods, fix long line
f7032eb [Holden Karau] Fix a java test bug, remove some unecessary changes
adf15b4 [Holden Karau] rename the classifier suite test to ProbabilisticClassifierSuite now that we only have it in Probabilistic
398078a [Holden Karau] move the thresholding around a bunch based on the design doc
4893bdc [Holden Karau] Use numtrees of 3 since previous result was tied (one tree for each) and the switch from different max methods picked a different element (since they were equal I think this is ok)
638854c [Holden Karau] Add a scala RandomForestClassifierSuite test based on corresponding python test
e09919c [Holden Karau] Fix return type, I need more coffee....
8d92cac [Holden Karau] Use ClassifierParams as the head
3456ed3 [Holden Karau] Add explicit return types even though just test
a0f3b0c [Holden Karau] scala style fixes
6f14314 [Holden Karau] Since hasthreshold/hasthresholds is in root classifier now
ffc8dab [Holden Karau] Update the sharedParams
0420290 [Holden Karau] Allow us to override the get methods selectively
978e77a [Holden Karau] Move HasThreshold into classifier params and start defining the overloaded getThreshold/getThresholds functions
1433e52 [Holden Karau] Revert "try and hide threshold but chainges the API so no dice there"
1f09a2e [Holden Karau] try and hide threshold but chainges the API so no dice there
efb9084 [Holden Karau] move setThresholds only to where its used
6b34809 [Holden Karau] Add a test with thresholding for the RFCS
74f54c3 [Holden Karau] Fix creation of vote array
1986fa8 [Holden Karau] Setting the thresholds only makes sense if the underlying class hasn't overridden predict, so lets push it down.
2f44b18 [Holden Karau] Add a global default of null for thresholds param
f338cfc [Holden Karau] Wait that wasn't a good idea, Revert "Some progress towards unifying threshold and thresholds"
634b06f [Holden Karau] Some progress towards unifying threshold and thresholds
85c9e01 [Holden Karau] Test passes again... little fnur
099c0f3 [Holden Karau] Move thresholds around some more (set on model not trainer)
0f46836 [Holden Karau] Start adding a classifiersuite
f70eb5e [Holden Karau] Fix test compile issues
a7d59c8 [Holden Karau] Move thresholding into Classifier trait
5d999d2 [Holden Karau] Some more progress, start adding a test (maybe try and see if we can find a better thing to use for the base of the test)
1fed644 [Holden Karau] Use thresholds to scale scores in random forest classifcation
31d6bf2 [Holden Karau] Start threading the threshold info through
0ef228c [Holden Karau] Add hasthresholds
Enable most javac lint warnings; fix a lot of build warnings. In a few cases, touch up surrounding code in the process.
I'll explain several of the changes inline in comments.
Author: Sean Owen <sowen@cloudera.com>
Closes#7862 from srowen/SPARK-9534 and squashes the following commits:
ea51618 [Sean Owen] Enable most javac lint warnings; fix a lot of build warnings. In a few cases, touch up surrounding code in the process.
Add missing methods
1. getVectors
2. findSynonyms
to W2Vec scala and python API
mengxr
Author: MechCoder <manojkumarsivaraj334@gmail.com>
Closes#7263 from MechCoder/missing_methods_w2vec and squashes the following commits:
149d5ca [MechCoder] minor doc
69d91b7 [MechCoder] [SPARK-8874] [ML] Add missing methods in Word2Vec
Add Python API for RFormula. Similar to other feature transformers in Python. This is just a thin wrapper over the Scala implementation. ericl MechCoder
Author: Xiangrui Meng <meng@databricks.com>
Closes#7879 from mengxr/SPARK-9544 and squashes the following commits:
3d5ff03 [Xiangrui Meng] add an doctest for . and -
5e969a5 [Xiangrui Meng] fix pydoc
1cd41f8 [Xiangrui Meng] organize imports
3c18b10 [Xiangrui Meng] add Python API for RFormula
Added featureImportance to RandomForestClassifier and Regressor.
This follows the scikit-learn implementation here: [a95203b249/sklearn/tree/_tree.pyx (L3341)]
CC: yanboliang Would you mind taking a look? Thanks!
Author: Joseph K. Bradley <joseph@databricks.com>
Author: Feynman Liang <fliang@databricks.com>
Closes#7838 from jkbradley/dt-feature-importance and squashes the following commits:
72a167a [Joseph K. Bradley] fixed unit test
86cea5f [Joseph K. Bradley] Modified RF featuresImportances to return Vector instead of Map
5aa74f0 [Joseph K. Bradley] finally fixed unit test for real
33df5db [Joseph K. Bradley] fix unit test
42a2d3b [Joseph K. Bradley] fix unit test
fe94e72 [Joseph K. Bradley] modified feature importance unit tests
cc693ee [Feynman Liang] Add classifier tests
79a6f87 [Feynman Liang] Compare dense vectors in test
21d01fc [Feynman Liang] Added failing SKLearn test
ac0b254 [Joseph K. Bradley] Added featureImportance to RandomForestClassifier/Regressor. Need to add unit tests
RandomForestClassifier now outputs rawPrediction based on tree probabilities, plus probability column computed from normalized rawPrediction.
CC: holdenk
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#7859 from jkbradley/rf-prob and squashes the following commits:
6c28f51 [Joseph K. Bradley] Changed RandomForestClassifier to extend ProbabilisticClassifier
1. Use `PrefixSpanModel` to wrap the frequent sequences.
2. Define `FreqSequence` to wrap each frequent sequence, which contains a Java-friendly method `javaSequence`
3. Overload `run` for Java users.
4. Added a unit test in Java to check Java compatibility.
zhangjiajin feynmanliang
Author: Xiangrui Meng <meng@databricks.com>
Closes#7869 from mengxr/SPARK-9527 and squashes the following commits:
4345594 [Xiangrui Meng] add PrefixSpanModel and make PrefixSpan Java friendly
mengxr Please review after #7818 merges and master is rebased.
Continues work by rikima
Closes#7400
Author: Feynman Liang <fliang@databricks.com>
Author: masaki rikitoku <rikima3132@gmail.com>
Closes#7837 from feynmanliang/SPARK-7400-genericItems and squashes the following commits:
8b2c756 [Feynman Liang] Remove orig
92443c8 [Feynman Liang] Style fixes
42c6349 [Feynman Liang] Style fix
14e67fc [Feynman Liang] Generic prefixSpan itemtypes
b3b21e0 [Feynman Liang] Initial support for generic itemtype in public api
b86e0d5 [masaki rikitoku] modify to support generic item type
Remove ScalaDoc that suggests describeTopics and topDocumentsPerTopic are approximate.
cc jkbradley
Author: Meihua Wu <meihuawu@umich.edu>
Closes#7858 from rotationsymmetry/SPARK-9530 and squashes the following commits:
b574923 [Meihua Wu] Remove ScalaDoc that suggests describeTopics and topDocumentsPerTopic are approximate.
jira: https://issues.apache.org/jira/browse/SPARK-8169
stop words: http://en.wikipedia.org/wiki/Stop_words
StopWordsRemover takes a string array column and outputs a string array column with all defined stop words removed. The transformer should also come with a standard set of stop words as default.
Currently I used a minimum stop words set since on some [case](http://nlp.stanford.edu/IR-book/html/htmledition/dropping-common-terms-stop-words-1.html), small set of stop words is preferred.
ASCII char has been tested, Yet I cannot check it in due to style check.
Further thought,
1. Maybe I should use OpenHashSet. Is it recommended?
2. Currently I leave the null in input array untouched, i.e. Array(null, null) => Array(null, null).
3. If the current stop words set looks too limited, any suggestion for replacement? We can have something similar to the one in [SKlearn](https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/feature_extraction/stop_words.py).
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#6742 from hhbyyh/stopwords and squashes the following commits:
fa959d8 [Yuhao Yang] separating udf
f190217 [Yuhao Yang] replace default list and other small fix
04403ab [Yuhao Yang] Merge remote-tracking branch 'upstream/master' into stopwords
b3aa957 [Yuhao Yang] add stopWordsRemover
mengxr Extends PrefixSpan to non-temporal itemsets. Continues work by zhangjiajin
* Internal API uses List[Set[Int]] which is likely not efficient; will need to refactor during QA
Closes#7646
Author: zhangjiajin <zhangjiajin@huawei.com>
Author: Feynman Liang <fliang@databricks.com>
Author: zhang jiajin <zhangjiajin@huawei.com>
Closes#7818 from feynmanliang/SPARK-8999-nonTemporal and squashes the following commits:
4ded81d [Feynman Liang] Replace all filters to filter nonempty
350e67e [Feynman Liang] Code review feedback
03156ca [Feynman Liang] Fix tests, drop delimiters at boundaries of sequences
d1fe0ed [Feynman Liang] Remove comments
86ca4e5 [Feynman Liang] Fix style
7c7bf39 [Feynman Liang] Fixed itemSet sequences
6073b10 [Feynman Liang] Basic itemset functionality, failing test
1a7fb48 [Feynman Liang] Add delimiter to results
5db00aa [Feynman Liang] Working for items, not itemsets
6787716 [Feynman Liang] Working on temporal sequences
f1114b9 [Feynman Liang] Add -1 delimiter
00fe756 [Feynman Liang] Reset base files for rebase
f486dcd [zhangjiajin] change maxLocalProjDBSize and fix a bug (remove -3 from frequent items).
60a0b76 [zhangjiajin] fixed a scala style error.
740c203 [zhangjiajin] fixed a scala style error.
5785cb8 [zhangjiajin] support non-temporal sequence
a5d649d [zhangjiajin] restore original version
09dc409 [zhangjiajin] Merge branch 'master' of https://github.com/apache/spark into multiItems_2
ae8c02d [zhangjiajin] Fixed some Scala style errors.
216ab0c [zhangjiajin] Support non-temporal sequence in PrefixSpan
b572f54 [zhangjiajin] initialize file before rebase.
f06772f [zhangjiajin] fix a scala style error.
a7e50d4 [zhangjiajin] Add feature: Collect enough frequent prefixes before projection in PrefixSpan.
c1d13d0 [zhang jiajin] Delete PrefixspanSuite.scala
d9d8137 [zhang jiajin] Delete Prefixspan.scala
c6ceb63 [zhangjiajin] Add new algorithm PrefixSpan and test file.
It is useful to convert the encoded indices back to their string representation for result inspection. We can add a function which creates an inverse transformation.
Author: Holden Karau <holden@pigscanfly.ca>
Closes#6339 from holdenk/SPARK-7446-inverse-transform-for-string-indexer and squashes the following commits:
7cdf915 [Holden Karau] scala style comment fix
b9cffb6 [Holden Karau] Update the labels param to have the metadata note
6a38edb [Holden Karau] Setting the default needs to come after the value gets defined
9e241d8 [Holden Karau] use Array.empty
21c8cfa [Holden Karau] Merge branch 'master' into SPARK-7446-inverse-transform-for-string-indexer
64dd3a3 [Holden Karau] Merge branch 'master' into SPARK-7446-inverse-transform-for-string-indexer
4f06c59 [Holden Karau] Fix comment styles, use empty array as the default, etc.
a60c0e3 [Holden Karau] CR feedback (remove old constructor, add a note about use of setLabels)
1987b95 [Holden Karau] Use default copy
71e8d66 [Holden Karau] Make labels a local param for StringIndexerInverse
8450d0b [Holden Karau] Use the labels param in StringIndexerInverse
7464019 [Holden Karau] Add a labels param
868b1a9 [Holden Karau] Update scaladoc since we don't have labelsCol anymore
5aa38bf [Holden Karau] Add an inverse test using only meta data, pass labels when calling inverse method
f3e0c64 [Holden Karau] CR feedback
ebed932 [Holden Karau] Add Experimental tag and some scaladocs. Also don't require that the inputCol has the metadata on it, instead have the labelsCol specified when creating the inverse.
03ebf95 [Holden Karau] Add explicit type for invert function
ecc65e0 [Holden Karau] Read the metadata correctly, use the array, pass the test
a42d773 [Holden Karau] Fix test to supply cols as per new invert method
16cc3c3 [Holden Karau] Add an invert method
d4bcb20 [Holden Karau] Make the inverse string indexer into a transformer (still needs test updates but compiles)
e8bf3ad [Holden Karau] Merge branch 'master' into SPARK-7446-inverse-transform-for-string-indexer
c3fdee1 [Holden Karau] Some WIP refactoring based on jkbradley's CR feedback. Definite work-in-progress
557bef8 [Holden Karau] Instead of using a private inverse transform, add an invert function so we can use it in a pipeline
88779c1 [Holden Karau] fix long line
78b28c1 [Holden Karau] Finish reverse part and add a test :)
bb16a6a [Holden Karau] Some progress
This PR adds a `MapData` as internal representation of map type in Spark SQL, and provides a default implementation with just 2 `ArrayData`.
After that, we have specialized getters for all internal type, so I removed generic getter in `ArrayData` and added specialized `toArray` for it.
Also did some refactor and cleanup for `InternalRow` and its subclasses.
Author: Wenchen Fan <cloud0fan@outlook.com>
Closes#7799 from cloud-fan/map-data and squashes the following commits:
77d482f [Wenchen Fan] fix python
e8f6682 [Wenchen Fan] skip MapData equality check in HiveInspectorSuite
40cc9db [Wenchen Fan] add toString
6e06ec9 [Wenchen Fan] some more cleanup
a90aca1 [Wenchen Fan] add MapData
Adds `alpha` (document-topic Dirichlet parameter) hyperparameter optimization to `OnlineLDAOptimizer` following Huang: Maximum Likelihood Estimation of Dirichlet Distribution Parameters. Also introduces a private `setSampleWithReplacement` to `OnlineLDAOptimizer` for unit testing purposes.
Author: Feynman Liang <fliang@databricks.com>
Closes#7836 from feynmanliang/SPARK-8936-alpha-optimize and squashes the following commits:
4bef484 [Feynman Liang] Documentation improvements
c3c6c1d [Feynman Liang] Fix docs
151e859 [Feynman Liang] Fix style
fa77518 [Feynman Liang] Hyperparameter optimization
Make NaiveBayesModel support predicting class probabilities, inherit from ProbabilisticClassificationModel.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#7672 from yanboliang/spark-9308 and squashes the following commits:
25e224c [Yanbo Liang] raw2probabilityInPlace should operate in-place
3ee56d6 [Yanbo Liang] change predictRaw and raw2probabilityInPlace
c07e7a2 [Yanbo Liang] ml.NaiveBayesModel support predicting class probabilities
Add topDocumentsPerTopic to DistributedLDAModel.
Add ScalaDoc and unit tests.
Author: Meihua Wu <meihuawu@umich.edu>
Closes#7769 from rotationsymmetry/SPARK-9246 and squashes the following commits:
1029e79c [Meihua Wu] clean up code comments
a023b82 [Meihua Wu] Update tests to use Long for doc index.
91e5998 [Meihua Wu] Use Long for doc index.
b9f70cf [Meihua Wu] Revise topDocumentsPerTopic
26ff3f6 [Meihua Wu] Add topDocumentsPerTopic, scala doc and unit tests
jkbradley Exposes `bound` (variational log likelihood bound) through public API as `logLikelihood`. Also adds unit tests, some DRYing of `LDASuite`, and includes unit tests mentioned in #7760
Author: Feynman Liang <fliang@databricks.com>
Closes#7801 from feynmanliang/SPARK-9481-logLikelihood and squashes the following commits:
6d1b2c9 [Feynman Liang] Negate perplexity definition
5f62b20 [Feynman Liang] Add logLikelihood
Decision tree support predict class probabilities.
Implement the prediction probabilities function referred the old DecisionTree API and the [sklean API](https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/tree/tree.py#L593).
I make the DecisionTreeClassificationModel inherit from ProbabilisticClassificationModel, make the predictRaw to return the raw counts vector and make raw2probabilityInPlace/predictProbability return the probabilities for each prediction.
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#7694 from yanboliang/spark-6885 and squashes the following commits:
08d5b7f [Yanbo Liang] fix ImpurityStats null parameters and raw2probabilityInPlace sum = 0 issue
2174278 [Yanbo Liang] solve merge conflicts
7e90ba8 [Yanbo Liang] fix typos
33ae183 [Yanbo Liang] fix annotation
ff043d3 [Yanbo Liang] raw2probabilityInPlace should operate in-place
c32d6ce [Yanbo Liang] optimize calculateImpurityStats function again
6167fb0 [Yanbo Liang] optimize calculateImpurityStats function
fbbe2ec [Yanbo Liang] eliminate duplicated struct and code
beb1634 [Yanbo Liang] try to eliminate impurityStats for each LearningNode
99e8943 [Yanbo Liang] code optimization
5ec3323 [Yanbo Liang] implement InformationGainAndImpurityStats
227c91b [Yanbo Liang] refactor LearningNode to store ImpurityCalculator
d746ffc [Yanbo Liang] decision tree support predict class probabilities
jira: https://issues.apache.org/jira/browse/SPARK-9231
Helper method in DistributedLDAModel of this form:
```
/**
* For each document, return the top k weighted topics for that document.
* return RDD of (doc ID, topic indices, topic weights)
*/
def topTopicsPerDocument(k: Int): RDD[(Long, Array[Int], Array[Double])]
```
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#7785 from hhbyyh/topTopicsPerdoc and squashes the following commits:
30ad153 [Yuhao Yang] small fix
fd24580 [Yuhao Yang] add topTopics per document to DistributedLDAModel
This pull request contains the following feature for ML:
- Multilayer Perceptron classifier
This implementation is based on our initial pull request with bgreeven: https://github.com/apache/spark/pull/1290 and inspired by very insightful suggestions from mengxr and witgo (I would like to thank all other people from the mentioned thread for useful discussions). The original code was extensively tested and benchmarked. Since then, I've addressed two main requirements that prevented the code from merging into the main branch:
- Extensible interface, so it will be easy to implement new types of networks
- Main building blocks are traits `Layer` and `LayerModel`. They are used for constructing layers of ANN. New layers can be added by extending the `Layer` and `LayerModel` traits. These traits are private in this release in order to save path to improve them based on community feedback
- Back propagation is implemented in general form, so there is no need to change it (optimization algorithm) when new layers are implemented
- Speed and scalability: this implementation has to be comparable in terms of speed to the state of the art single node implementations.
- The developed benchmark for large ANN shows that the proposed code is on par with C++ CPU implementation and scales nicely with the number of workers. Details can be found here: https://github.com/avulanov/ann-benchmark
- DBN and RBM by witgo https://github.com/witgo/spark/tree/ann-interface-gemm-dbn
- Dropout https://github.com/avulanov/spark/tree/ann-interface-gemm
mengxr and dbtsai kindly agreed to perform code review.
Author: Alexander Ulanov <nashb@yandex.ru>
Author: Bert Greevenbosch <opensrc@bertgreevenbosch.nl>
Closes#7621 from avulanov/SPARK-2352-ann and squashes the following commits:
4806b6f [Alexander Ulanov] Addressing reviewers comments.
a7e7951 [Alexander Ulanov] Default blockSize: 100. Added documentation to blockSize parameter and DataStacker class
f69bb3d [Alexander Ulanov] Addressing reviewers comments.
374bea6 [Alexander Ulanov] Moving ANN to ML package. GradientDescent constructor is now spark private.
43b0ae2 [Alexander Ulanov] Addressing reviewers comments. Adding multiclass test.
9d18469 [Alexander Ulanov] Addressing reviewers comments: unnecessary copy of data in predict
35125ab [Alexander Ulanov] Style fix in tests
e191301 [Alexander Ulanov] Apache header
a226133 [Alexander Ulanov] Multilayer Perceptron regressor and classifier
support ml.NaiveBayes for Python
Author: Yanbo Liang <ybliang8@gmail.com>
Closes#7568 from yanboliang/spark-9214 and squashes the following commits:
5ee3fd6 [Yanbo Liang] fix typos
3ecd046 [Yanbo Liang] fix typos
f9c94d1 [Yanbo Liang] change lambda_ to smoothing and fix other issues
180452a [Yanbo Liang] fix typos
7dda1f4 [Yanbo Liang] support ml.NaiveBayes for Python
Improve error message when number of examples is less than arity of high-arity categorical feature
CC jkbradley is this about what you had in mind? I know it's a starter, but was on my list to close out in the short term.
Author: Sean Owen <sowen@cloudera.com>
Closes#7800 from srowen/SPARK-9077 and squashes the following commits:
b8f6cdb [Sean Owen] Improve error message when number of examples is less than arity of high-arity categorical feature
Add checkpointing to GradientBoostedTrees, GBTClassifier, GBTRegressor
CC: mengxr
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#7804 from jkbradley/gbt-checkpoint3 and squashes the following commits:
3fbd7ba [Joseph K. Bradley] tiny fix
b3e160c [Joseph K. Bradley] unset checkpoint dir after test
9cc3a04 [Joseph K. Bradley] added checkpointing to GBTs
Author: martinzapletal <zapletal-martin@email.cz>
Closes#7517 from zapletal-martin/SPARK-8671-isotonic-regression-api and squashes the following commits:
8c435c1 [martinzapletal] Review https://github.com/apache/spark/pull/7517 feedback update.
bebbb86 [martinzapletal] Merge remote-tracking branch 'upstream/master' into SPARK-8671-isotonic-regression-api
b68efc0 [martinzapletal] Added tests for param validation.
07c12bd [martinzapletal] Comments and refactoring.
834fcf7 [martinzapletal] Merge remote-tracking branch 'upstream/master' into SPARK-8671-isotonic-regression-api
b611fee [martinzapletal] SPARK-8671. Added first version of isotonic regression to pipeline API
See https://issues.apache.org/jira/browse/SPARK-9479 for the failure cause.
The PR includes the following changes:
1. Make ReceiverTrackerSuite create StreamingContext in the test body.
2. Fix places that don't stop StreamingContext. I verified no SparkContext was stopped in the shutdown hook locally after this fix.
3. Fix an issue that `ReceiverTracker.endpoint` may be null.
4. Make sure stopping SparkContext in non-main thread won't fail other tests.
Author: zsxwing <zsxwing@gmail.com>
Closes#7797 from zsxwing/fix-ReceiverTrackerSuite and squashes the following commits:
3a4bb98 [zsxwing] Fix another potential NPE
d7497df [zsxwing] Fix ReceiverTrackerSuite; make sure StreamingContext in tests is closed
jkbradley Changes the current hacky string-comparison for vector compares.
Author: Feynman Liang <fliang@databricks.com>
Closes#7775 from feynmanliang/SPARK-9454-ldasuite-vector-compare and squashes the following commits:
bd91a82 [Feynman Liang] Remove println
905c76e [Feynman Liang] Fix string compare in distributed EM
2f24c13 [Feynman Liang] Improve LDASuite tests
Check that SparseVector size is at least as big as the number of indices/values provided. And add tests for constructor checks.
CC MechCoder jkbradley -- I am not sure if a change needs to also happen in the Python API? I didn't see it had any similar checks to begin with, but I don't know it well.
Author: Sean Owen <sowen@cloudera.com>
Closes#7794 from srowen/SPARK-9277 and squashes the following commits:
e8dc31e [Sean Owen] Fix scalastyle
6ffe34a [Sean Owen] Check that SparseVector size is at least as big as the number of indices/values provided. And add tests for constructor checks.
Add unit tests for running LDA with empty documents.
Both EMLDAOptimizer and OnlineLDAOptimizer are tested.
feynmanliang
Author: Meihua Wu <meihuawu@umich.edu>
Closes#7620 from rotationsymmetry/SPARK-9225 and squashes the following commits:
3ed7c88 [Meihua Wu] Incorporate reviewer's further comments
f9432e8 [Meihua Wu] Incorporate reviewer's comments
8e1b9ec [Meihua Wu] Merge remote-tracking branch 'upstream/master' into SPARK-9225
ad55665 [Meihua Wu] Add unit tests for running LDA with empty documents
A trivial fix for the comments of RegexTokenizer.
Maybe this is too small, yet I just noticed it and think it can be quite misleading. I can create a jira if necessary.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#7791 from hhbyyh/docFix and squashes the following commits:
cdf2542 [Yuhao Yang] minor fix on tokenizer doc
Continuation of work by zhangjiajin
Closes#7412
Author: zhangjiajin <zhangjiajin@huawei.com>
Author: Feynman Liang <fliang@databricks.com>
Author: zhang jiajin <zhangjiajin@huawei.com>
Closes#7783 from feynmanliang/SPARK-8998-improve-distributed and squashes the following commits:
a61943d [Feynman Liang] Collect small patterns to local
4ddf479 [Feynman Liang] Parallelize freqItemCounts
ad23aa9 [zhang jiajin] Merge pull request #1 from feynmanliang/SPARK-8998-collectBeforeLocal
87fa021 [Feynman Liang] Improve extend prefix readability
c2caa5c [Feynman Liang] Readability improvements and comments
1235cfc [Feynman Liang] Use Iterable[Array[_]] over Array[Array[_]] for database
da0091b [Feynman Liang] Use lists for prefixes to reuse data
cb2a4fc [Feynman Liang] Inline code for readability
01c9ae9 [Feynman Liang] Add getters
6e149fa [Feynman Liang] Fix splitPrefixSuffixPairs
64271b3 [zhangjiajin] Modified codes according to comments.
d2250b7 [zhangjiajin] remove minPatternsBeforeLocalProcessing, add maxSuffixesBeforeLocalProcessing.
b07e20c [zhangjiajin] Merge branch 'master' of https://github.com/apache/spark into CollectEnoughPrefixes
095aa3a [zhangjiajin] Modified the code according to the review comments.
baa2885 [zhangjiajin] Modified the code according to the review comments.
6560c69 [zhangjiajin] Add feature: Collect enough frequent prefixes before projection in PrefixeSpan
a8fde87 [zhangjiajin] Merge branch 'master' of https://github.com/apache/spark
4dd1c8a [zhangjiajin] initialize file before rebase.
078d410 [zhangjiajin] fix a scala style error.
22b0ef4 [zhangjiajin] Add feature: Collect enough frequent prefixes before projection in PrefixSpan.
ca9c4c8 [zhangjiajin] Modified the code according to the review comments.
574e56c [zhangjiajin] Add new object LocalPrefixSpan, and do some optimization.
ba5df34 [zhangjiajin] Fix a Scala style error.
4c60fb3 [zhangjiajin] Fix some Scala style errors.
1dd33ad [zhangjiajin] Modified the code according to the review comments.
89bc368 [zhangjiajin] Fixed a Scala style error.
a2eb14c [zhang jiajin] Delete PrefixspanSuite.scala
951fd42 [zhang jiajin] Delete Prefixspan.scala
575995f [zhangjiajin] Modified the code according to the review comments.
91fd7e6 [zhangjiajin] Add new algorithm PrefixSpan and test file.
PeriodicGraphCheckpointer was introduced for Latent Dirichlet Allocation (LDA), but it was meant to be generalized to work with Graphs, RDDs, and other data structures based on RDDs. This PR generalizes it.
For those who are not familiar with the periodic checkpointer, it tries to automatically handle persisting/unpersisting and checkpointing/removing checkpoint files in a lineage of RDD-based objects.
I need it generalized to use with GradientBoostedTrees [https://issues.apache.org/jira/browse/SPARK-6684]. It should be useful for other iterative algorithms as well.
Changes I made:
* Copied PeriodicGraphCheckpointer to PeriodicCheckpointer.
* Within PeriodicCheckpointer, I created abstract methods for the basic operations (checkpoint, persist, etc.).
* The subclasses for Graphs and RDDs implement those abstract methods.
* I copied the test suite for the graph checkpointer and made tiny modifications to make it work for RDDs.
To review this PR, I recommend doing 2 diffs:
(1) diff between the old PeriodicGraphCheckpointer.scala and the new PeriodicCheckpointer.scala
(2) diff between the 2 test suites
CCing andrewor14 in case there are relevant changes to checkpointing.
CCing feynmanliang in case you're interested in learning about checkpointing.
CCing mengxr for final OK.
Thanks all!
Author: Joseph K. Bradley <joseph@databricks.com>
Closes#7728 from jkbradley/gbt-checkpoint and squashes the following commits:
d41902c [Joseph K. Bradley] Oops, forgot to update an extra time in the checkpointer tests, after the last commit. I'll fix that. I'll also make some of the checkpointer methods protected, which I should have done before.
32b23b8 [Joseph K. Bradley] fixed usage of checkpointer in lda
0b3dbc0 [Joseph K. Bradley] Changed checkpointer constructor not to take initial data.
568918c [Joseph K. Bradley] Generalized PeriodicGraphCheckpointer to PeriodicCheckpointer, with subclasses for RDDs and Graphs.
jira: https://issues.apache.org/jira/browse/SPARK-7368
Add QR decomposition for RowMatrix.
I'm not sure what's the blueprint about the distributed Matrix from community and whether this will be a desirable feature , so I sent a prototype for discussion. I'll go on polish the code and provide ut and performance statistics if it's acceptable.
The implementation refers to the [paper: https://www.cs.purdue.edu/homes/dgleich/publications/Benson%202013%20-%20direct-tsqr.pdf]
Austin R. Benson, David F. Gleich, James Demmel. "Direct QR factorizations for tall-and-skinny matrices in MapReduce architectures", 2013 IEEE International Conference on Big Data, which is a stable algorithm with good scalability.
Currently I tried it on a 400000 * 500 rowMatrix (16 partitions) and it can bring down the computation time from 8.8 mins (using breeze.linalg.qr.reduced) to 2.6 mins on a 4 worker cluster. I think there will still be some room for performance improvement.
Any trial and suggestion is welcome.
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes#5909 from hhbyyh/qrDecomposition and squashes the following commits:
cec797b [Yuhao Yang] remove unnecessary qr
0fb1012 [Yuhao Yang] hierarchy R computing
3fbdb61 [Yuhao Yang] update qr to indirect and add ut
0d913d3 [Yuhao Yang] Merge remote-tracking branch 'upstream/master' into qrDecomposition
39213c3 [Yuhao Yang] Merge remote-tracking branch 'upstream/master' into qrDecomposition
c0fc0c7 [Yuhao Yang] Merge remote-tracking branch 'upstream/master' into qrDecomposition
39b0b22 [Yuhao Yang] initial draft for discussion
jkbradley MechCoder
Resolves blocking issue for SPARK-6793. Please review after #7705 is merged.
Author: Feynman Liang <fliang@databricks.com>
Closes#7757 from feynmanliang/SPARK-9940-localSaveLoad and squashes the following commits:
d0d8cf4 [Feynman Liang] Fix thisClassName
0f30109 [Feynman Liang] Fix tests after changing LDAModel public API
dc61981 [Feynman Liang] Add hyperparams to LocalLDAModel save/load
Implement the classification trait for RandomForestClassifiers. The plan is to use this in the future to providing thresholding for RandomForestClassifiers (as well as other classifiers that implement that trait).
Author: Holden Karau <holden@pigscanfly.ca>
Closes#7432 from holdenk/SPARK-9016-make-random-forest-classifiers-implement-classification-trait and squashes the following commits:
bf22fa6 [Holden Karau] Add missing imports for testing suite
e948f0d [Holden Karau] Check the prediction generation from rawprediciton
25320c3 [Holden Karau] Don't supply numClasses when not needed, assert model classes are as expected
1a67e04 [Holden Karau] Use old decission tree stuff instead
673e0c3 [Holden Karau] Merge branch 'master' into SPARK-9016-make-random-forest-classifiers-implement-classification-trait
0d15b96 [Holden Karau] FIx typo
5eafad4 [Holden Karau] add a constructor for rootnode + num classes
fc6156f [Holden Karau] scala style fix
2597915 [Holden Karau] take num classes in constructor
3ccfe4a [Holden Karau] Merge in master, make pass numClasses through randomforest for training
222a10b [Holden Karau] Increase numtrees to 3 in the python test since before the two were equal and the argmax was selecting the last one
16aea1c [Holden Karau] Make tests match the new models
b454a02 [Holden Karau] Make the Tree classifiers extends the Classifier base class
77b4114 [Holden Karau] Import vectors lib
Author: Bimal Tandel <bimal@bimal-MBP.local>
Closes#7730 from BimalTandel/branch_spark_8921 and squashes the following commits:
3ea230a [Bimal Tandel] Spark 8921 add @since tags
This PR introduces save / load for GMM's in python API.
Also I refactored `GaussianMixtureModel` and inherited it from `JavaModelWrapper` with model being `GaussianMixtureModelWrapper`, a wrapper which provides convenience methods to `GaussianMixtureModel` (due to serialization and deserialization issues) and I moved the creation of gaussians to the scala backend.
Author: MechCoder <manojkumarsivaraj334@gmail.com>
Closes#7617 from MechCoder/python_gmm_save_load and squashes the following commits:
9c305aa [MechCoder] [SPARK-7105] [PySpark] [MLlib] Support model save/load in GMM
Author: vinodkc <vinod.kc.in@gmail.com>
Closes#7325 from vinodkc/add_since_mllib.recommendation and squashes the following commits:
93156f2 [vinodkc] Changed 0.8.0 to 0.9.1
c413350 [vinodkc] Added @since
This adds StringType feature support via OneHotEncoder. As part of this task it was necessary to change RFormula to an Estimator, so that factor levels could be determined from the training dataset.
Not sure if I am using uids correctly here, would be good to get reviewer help on that.
cc mengxr
Umbrella design doc: https://docs.google.com/document/d/10NZNSEurN2EdWM31uFYsgayIPfCFHiuIu3pCWrUmP_c/edit#
Author: Eric Liang <ekl@databricks.com>
Closes#7574 from ericl/string-features and squashes the following commits:
f99131a [Eric Liang] comments
0bf3c26 [Eric Liang] update docs
c302a2c [Eric Liang] fix tests
9d1ac82 [Eric Liang] Merge remote-tracking branch 'upstream/master' into string-features
e713da3 [Eric Liang] comments
4d79193 [Eric Liang] revert to seq + distinct
169a085 [Eric Liang] tweak functional test
a230a47 [Eric Liang] Merge branch 'master' into string-features
72bd6f3 [Eric Liang] fix merge
d841cec [Eric Liang] Merge branch 'master' into string-features
5b2c4a2 [Eric Liang] Mon Jul 20 18:45:33 PDT 2015
b01c7c5 [Eric Liang] add test
8a637db [Eric Liang] encoder wip
a1d03f4 [Eric Liang] refactor into estimator
Use Vector.argmax call instead of converting to dense vector before calculating predictions.
Author: George Dittmar <georgedittmar@gmail.com>
Closes#7670 from GeorgeDittmar/sprk-7423 and squashes the following commits:
e796747 [George Dittmar] Changing ClassificationModel and ProbabilisticClassificationModel to use Vector.argmax instead of converting to DenseVector
